]>
git.ipfire.org Git - thirdparty/wireguard-go.git/log
Jason A. Donenfeld [Tue, 23 Mar 2021 19:07:19 +0000 (13:07 -0600)]
version: bump snapshot
Jason A. Donenfeld [Tue, 23 Mar 2021 18:41:34 +0000 (12:41 -0600)]
tun: freebsd: use broadcast mode instead of PPP mode
It makes the routing configuration simpler.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Thu, 11 Mar 2021 16:29:10 +0000 (09:29 -0700)]
device: signal to close device in separate routine
Otherwise we wind up deadlocking.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Thu, 11 Mar 2021 16:23:11 +0000 (09:23 -0700)]
tun: linux: do not spam events every second from hack listener
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Kay Diam [Sun, 7 Mar 2021 16:21:31 +0000 (17:21 +0100)]
tun: freebsd: allow empty names
This change allows omitting the tun interface name setting. When the
name is not set, the kernel automatically picks up the tun name and
index.
Signed-off-by: Kay Diam <kay.diam@gmail.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 3 Mar 2021 11:26:59 +0000 (12:26 +0100)]
winpipe: move syscalls into x/sys
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 3 Mar 2021 14:05:19 +0000 (15:05 +0100)]
memmod: use resource functions from x/sys
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 3 Mar 2021 13:38:26 +0000 (14:38 +0100)]
memmod: do not use IsBadReadPtr
It should be enough to check for the trailing zero name.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Sat, 6 Mar 2021 16:20:46 +0000 (09:20 -0700)]
conn: linux: unexport mutex
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Fri, 5 Mar 2021 22:06:08 +0000 (15:06 -0700)]
mod: bump x/sys
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Sat, 6 Mar 2021 16:03:28 +0000 (09:03 -0700)]
mod: rename COPYING to LICENSE
Otherwise the netstack module doesn't show up on the package site.
https://github.com/golang/go/issues/43817#issuecomment-
764987580
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Sat, 6 Mar 2021 15:21:18 +0000 (08:21 -0700)]
tun/netstack: bump deps and api
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Thu, 25 Feb 2021 11:28:53 +0000 (12:28 +0100)]
device: get rid of peers.empty boolean in timersActive
There's no way for len(peers)==0 when a current peer has
isRunning==false.
This requires some struct reshuffling so that the uint64 pointer is
aligned.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Mon, 22 Feb 2021 17:47:41 +0000 (18:47 +0100)]
conn: implement RIO for fast Windows UDP sockets
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Mon, 22 Feb 2021 14:43:08 +0000 (15:43 +0100)]
global: remove TODO name graffiti
Googlers have a habit of graffiting their name in TODO items that then
are never addressed, and other people won't go near those because
they're marked territory of another animal. I've been gradually cleaning
these up as I see them, but this commit just goes all the way and
removes the remaining stragglers.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Mon, 22 Feb 2021 03:30:31 +0000 (04:30 +0100)]
device: test up/down using virtual conn
This prevents port clashing bugs.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Mon, 22 Feb 2021 01:57:41 +0000 (02:57 +0100)]
device: cleanup unused test components
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Mon, 22 Feb 2021 01:01:50 +0000 (02:01 +0100)]
conn: make binds replacable
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Mon, 22 Feb 2021 14:12:03 +0000 (15:12 +0100)]
device: disable waitpool tests
This code is stable, and the test is finicky, especially on high core
count systems, so just disable it.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Brad Fitzpatrick [Thu, 18 Feb 2021 22:53:22 +0000 (14:53 -0800)]
tun: make NativeTun.Close well behaved, not crash on double close
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Brad Fitzpatrick [Thu, 18 Feb 2021 22:42:04 +0000 (14:42 -0800)]
README: bump document Go requirement to 1.16
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Jason A. Donenfeld [Wed, 17 Feb 2021 21:19:27 +0000 (22:19 +0100)]
global: stop using ioutil
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 16 Feb 2021 20:05:25 +0000 (21:05 +0100)]
conn: bump to 1.16 and get rid of NetErrClosed hack
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Fri, 12 Feb 2021 17:00:59 +0000 (18:00 +0100)]
version: bump snapshot
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Fri, 12 Feb 2021 16:59:50 +0000 (17:59 +0100)]
device: remove old version file
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Thu, 11 Feb 2021 14:48:56 +0000 (15:48 +0100)]
gitignore: remove old hacks
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 10 Feb 2021 17:19:11 +0000 (18:19 +0100)]
device: use container/list instead of open coding it
This linked list implementation is awful, but maybe Go 2 will help
eventually, and at least we're not open coding the hlist any more.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 10 Feb 2021 00:01:37 +0000 (01:01 +0100)]
device: retry Up() in up/down test
We're loosing our ownership of the port when bringing the device down,
which means another test process could reclaim it. Avoid this by
retrying for 4 seconds.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 23:43:31 +0000 (00:43 +0100)]
conn: close old fd before trying again
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 23:39:28 +0000 (00:39 +0100)]
device: flush peer queues before starting device
In case some old packets snuck in there before, this flushes before
starting afresh.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 23:21:12 +0000 (00:21 +0100)]
device: create peer queues at peer creation time
Rather than racing with Start(), since we're never destroying these
queues, we just set the variables at creation time.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 23:12:23 +0000 (00:12 +0100)]
device: return error from Up() and Down()
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 19:18:21 +0000 (20:18 +0100)]
rwcancel: add an explicit close call
This lets us collect FDs even if the GC doesn't do it for us.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 18:54:00 +0000 (19:54 +0100)]
rwcancel: use errors.Is for unwrapping
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 18:48:27 +0000 (19:48 +0100)]
tun: use errors.Is for unwrapping
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 18:46:57 +0000 (19:46 +0100)]
conn: use errors.Is for unwrapping
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 18:26:45 +0000 (19:26 +0100)]
device: handshake routine writes into encryption queue
Since RoutineHandshake calls peer.SendKeepalive(), it potentially is a
writer into the encryption queue, so we need to bump the wg count.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Josh Bleecher Snyder [Tue, 9 Feb 2021 17:53:00 +0000 (09:53 -0800)]
device: make RoutineReadFromTUN keep encryption queue alive
RoutineReadFromTUN can trigger a call to SendStagedPackets.
SendStagedPackets attempts to protect against sending
on the encryption queue by checking peer.isRunning and device.isClosed.
However, those are subject to TOCTOU bugs.
If that happens, we get this:
goroutine 1254 [running]:
golang.zx2c4.com/wireguard/device.(*Peer).SendStagedPackets(0xc000798300)
.../wireguard-go/device/send.go:321 +0x125
golang.zx2c4.com/wireguard/device.(*Device).RoutineReadFromTUN(0xc000014780)
.../wireguard-go/device/send.go:271 +0x21c
created by golang.zx2c4.com/wireguard/device.NewDevice
.../wireguard-go/device/device.go:315 +0x298
Fix this with a simple, big hammer: Keep the encryption queue
alive as long as it might be written to.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 17:45:12 +0000 (18:45 +0100)]
conn: try harder to have v4 and v6 ports agree
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Josh Bleecher Snyder [Tue, 9 Feb 2021 17:08:17 +0000 (09:08 -0800)]
device: only allocate peer queues once
This serves two purposes.
First, it makes repeatedly stopping then starting a peer cheaper.
Second, it prevents a data race observed accessing the queues.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Tue, 9 Feb 2021 16:27:48 +0000 (08:27 -0800)]
device: clarify device.state.state docs (again)
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Tue, 9 Feb 2021 16:20:11 +0000 (08:20 -0800)]
device: run fewer iterations in TestUpDown
The high iteration count was useful when TestUpDown
was the nexus of new bugs to investigate.
Now that it has stabilized, that's less valuable.
And it slows down running the tests and crowds out other tests.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Tue, 9 Feb 2021 16:18:47 +0000 (08:18 -0800)]
device: run fewer trials in TestWaitPool when race detector enabled
On a many-core machine with the race detector enabled,
this test can take several minutes to complete.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Tue, 9 Feb 2021 16:15:21 +0000 (08:15 -0800)]
device: remove nil elem check in finalizers
This is not necessary, and removing it speeds up detection of UAF bugs.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 15:11:33 +0000 (16:11 +0100)]
device: rename unsafeRemovePeer to removePeerLocked
This matches the new naming scheme of upLocked and downLocked.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 14:39:19 +0000 (15:39 +0100)]
device: remove deviceStateNew
It's never used and we won't have a use for it. Also, move to go-running
stringer, for those without GOPATHs.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 14:35:43 +0000 (15:35 +0100)]
device: fix comment typo and shorten state.mu.Lock to state.Lock
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 14:32:55 +0000 (15:32 +0100)]
device: fix typo in comment
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 14:30:32 +0000 (15:30 +0100)]
device: fix alignment on 32-bit machines and test for it
The test previously checked the offset within a substruct, not the
offset within the allocated struct, so this adds the two together.
It then fixes an alignment crash on 32-bit machines.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 14:25:43 +0000 (15:25 +0100)]
device: do not log on idempotent device state change
Part of being actually idempotent is that we shouldn't penalize code
that takes advantage of this property with a log splat.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 14:09:50 +0000 (15:09 +0100)]
device: do not attach finalizer to non-returned object
Before, the code attached a finalizer to an object that wasn't returned,
resulting in immediate garbage collection. Instead return the actual
pointer.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 9 Feb 2021 14:00:59 +0000 (15:00 +0100)]
device: lock elem in autodraining queue before freeing
Without this, we wind up freeing packets that the encryption/decryption
queues still have, resulting in a UaF.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Mon, 8 Feb 2021 23:59:39 +0000 (00:59 +0100)]
device: remove listen port race in tests
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Mon, 8 Feb 2021 23:33:18 +0000 (00:33 +0100)]
device: generate test keys on the fly
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Josh Bleecher Snyder [Mon, 8 Feb 2021 21:02:52 +0000 (13:02 -0800)]
device: remove mutex from Peer send/receive
The immediate motivation for this change is an observed deadlock.
1. A goroutine calls peer.Stop. That calls peer.queue.Lock().
2. Another goroutine is in RoutineSequentialReceiver.
It receives an elem from peer.queue.inbound.
3. The peer.Stop goroutine calls close(peer.queue.inbound),
close(peer.queue.outbound), and peer.stopping.Wait().
It blocks waiting for RoutineSequentialReceiver
and RoutineSequentialSender to exit.
4. The RoutineSequentialReceiver goroutine calls peer.SendStagedPackets().
SendStagedPackets attempts peer.queue.RLock().
That blocks forever because the peer.Stop
goroutine holds a write lock on that mutex.
A background motivation for this change is that it can be expensive
to have a mutex in the hot code path of RoutineSequential*.
The mutex was necessary to avoid attempting to send elems on a closed channel.
This commit removes that danger by never closing the channel.
Instead, we send a sentinel nil value on the channel to indicate
to the receiver that it should exit.
The only problem with this is that if the receiver exits,
we could write an elem into the channel which would never get received.
If it never gets received, it cannot get returned to the device pools.
To work around this, we use a finalizer. When the channel can be GC'd,
the finalizer drains any remaining elements from the channel and
restores them to the device pool.
After that change, peer.queue.RWMutex no longer makes sense where it is.
It is only used to prevent concurrent calls to Start and Stop.
Move it to a more sensible location and make it a plain sync.Mutex.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Mon, 8 Feb 2021 20:38:19 +0000 (12:38 -0800)]
device: create channels.go
We have a bunch of stupid channel tricks, and I'm about to add more.
Give them their own file. This commit is 100% code movement.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Mon, 8 Feb 2021 19:36:55 +0000 (11:36 -0800)]
device: print direction when ping transit fails
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Mon, 8 Feb 2021 18:01:35 +0000 (10:01 -0800)]
device: separate timersInit from timersStart
timersInit sets up the timers.
It need only be done once per peer.
timersStart does the work to prepare the timers
for a newly running peer. It needs to be done
every time a peer starts.
Separate the two and call them in the appropriate places.
This prevents data races on the peer's timers fields
when starting and stopping peers.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Thu, 21 Jan 2021 17:26:14 +0000 (09:26 -0800)]
device: don't track device interface state in RoutineTUNEventReader
We already track this state elsewhere. No need to duplicate.
The cost of calling changeState is negligible.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Thu, 21 Jan 2021 17:23:45 +0000 (09:23 -0800)]
device: improve MTU change handling
The old code silently accepted negative MTUs.
It also set MTUs above the maximum.
It also had hard to follow deeply nested conditionals.
Add more paranoid handling,
and make the code more straight-line.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Mon, 8 Feb 2021 18:19:28 +0000 (10:19 -0800)]
device: remove device.state.stopping from RoutineTUNEventReader
The TUN event reader does three things: Change MTU, device up, and device down.
Changing the MTU after the device is closed does no harm.
Device up and device down don't make sense after the device is closed,
but we can check that condition before proceeding with changeState.
There's thus no reason to block device.Close on RoutineTUNEventReader exiting.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Tue, 19 Jan 2021 17:02:16 +0000 (09:02 -0800)]
device: overhaul device state management
This commit simplifies device state management.
It creates a single unified state variable and documents its semantics.
It also makes state changes more atomic.
As an example of the sort of bug that occurred due to non-atomic state changes,
the following sequence of events used to occur approximately every 2.5 million test runs:
* RoutineTUNEventReader received an EventDown event.
* It called device.Down, which called device.setUpDown.
* That set device.state.changing, but did not yet attempt to lock device.state.Mutex.
* Test completion called device.Close.
* device.Close locked device.state.Mutex.
* device.Close blocked on a call to device.state.stopping.Wait.
* device.setUpDown then attempted to lock device.state.Mutex and blocked.
Deadlock results. setUpDown cannot progress because device.state.Mutex is locked.
Until setUpDown returns, RoutineTUNEventReader cannot call device.state.stopping.Done.
Until device.state.stopping.Done gets called, device.state.stopping.Wait is blocked.
As long as device.state.stopping.Wait is blocked, device.state.Mutex cannot be unlocked.
This commit fixes that deadlock by holding device.state.mu
when checking that the device is not closed.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Mon, 8 Feb 2021 17:21:31 +0000 (09:21 -0800)]
device: remove unnecessary zeroing in peer.SendKeepalive
elem.packet is always already nil.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Wed, 3 Feb 2021 00:14:54 +0000 (16:14 -0800)]
device: remove device.state.stopping from RoutineHandshake
It is no longer necessary.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Tue, 19 Jan 2021 19:10:05 +0000 (11:10 -0800)]
device: remove device.state.stopping from RoutineDecryption
It is no longer necessary, as of
454de6f3e64abd2a7bf9201579cd92eea5280996
(device: use channel close to shut down and drain decryption channel).
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Jason A. Donenfeld [Thu, 4 Feb 2021 14:33:04 +0000 (15:33 +0100)]
main: add back version file
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 3 Feb 2021 16:56:46 +0000 (17:56 +0100)]
tai64n: add string representation for error messages
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 3 Feb 2021 16:52:31 +0000 (17:52 +0100)]
device: take peer handshake when reinitializing last sent handshake
This papers over other unrelated races, unfortunately.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Josh Bleecher Snyder [Wed, 3 Feb 2021 16:26:27 +0000 (08:26 -0800)]
device: fix goroutine leak test
The leak test had rare flakes.
If a system goroutine started at just the wrong moment, you'd get a false positive.
Instead of looping until the goroutines look good and then checking,
exit completely as soon as the number of goroutines looks good.
Also, check more frequently, in an attempt to complete faster.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Jason A. Donenfeld [Wed, 3 Feb 2021 16:43:41 +0000 (17:43 +0100)]
device: add up/down stress test
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 3 Feb 2021 16:29:01 +0000 (17:29 +0100)]
device: pass cfg strings around in tests instead of reader
This makes it easier to tag things onto the end manually for quick hacks.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 3 Feb 2021 15:54:45 +0000 (16:54 +0100)]
device: benchmark the waitpool to compare it to the prior channels
Here is the old implementation:
type WaitPool struct {
c chan interface{}
}
func NewWaitPool(max uint32, new func() interface{}) *WaitPool {
p := &WaitPool{c: make(chan interface{}, max)}
for i := uint32(0); i < max; i++ {
p.c <- new()
}
return p
}
func (p *WaitPool) Get() interface{} {
return <- p.c
}
func (p *WaitPool) Put(x interface{}) {
p.c <- x
}
It performs worse than the new one:
name old time/op new time/op delta
WaitPool-16 16.4µs ± 5% 15.1µs ± 3% -7.86% (p=0.008 n=5+5)
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Josh Bleecher Snyder [Tue, 2 Feb 2021 18:41:20 +0000 (10:41 -0800)]
device: test that we do not leak goroutines
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Tue, 2 Feb 2021 18:46:34 +0000 (10:46 -0800)]
device: tie encryption queue lifetime to the peers that write to it
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Jason A. Donenfeld [Tue, 2 Feb 2021 17:37:49 +0000 (18:37 +0100)]
device: use a waiting sync.Pool instead of a channel
Channels are FIFO which means we have guaranteed cache misses.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Fri, 29 Jan 2021 19:10:48 +0000 (20:10 +0100)]
device: reduce number of append calls when padding
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Fri, 29 Jan 2021 17:54:19 +0000 (18:54 +0100)]
device: use int64 instead of atomic.Value for time stamp
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Fri, 29 Jan 2021 17:24:45 +0000 (18:24 +0100)]
device: use new model queues for handshakes
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Fri, 29 Jan 2021 13:54:11 +0000 (14:54 +0100)]
device: simplify peer queue locking
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Thu, 28 Jan 2021 17:56:58 +0000 (18:56 +0100)]
device: reduce nesting when staging packet
Suggested-by: Josh Bleecher Snyder <josh@tailscale.com>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Thu, 28 Jan 2021 16:52:15 +0000 (17:52 +0100)]
global: bump copyright
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Thu, 28 Jan 2021 16:23:39 +0000 (17:23 +0100)]
device: remove version string
This is what modules are for, and Go binaries can introspect.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Thu, 28 Jan 2021 14:26:22 +0000 (15:26 +0100)]
device: do not allow get to run while set runs
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 27 Jan 2021 23:49:31 +0000 (00:49 +0100)]
device: avoid hex allocations in IpcGet
benchmark old ns/op new ns/op delta
BenchmarkUAPIGet-16 2872 2157 -24.90%
benchmark old allocs new allocs delta
BenchmarkUAPIGet-16 30 18 -40.00%
benchmark old bytes new bytes delta
BenchmarkUAPIGet-16 737 256 -65.26%
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Thu, 28 Jan 2021 13:44:51 +0000 (14:44 +0100)]
device: the psk is not a chapoly key
It's a separate type of key that gets hashed into the chain.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 27 Jan 2021 17:13:53 +0000 (18:13 +0100)]
device: get rid of nonce routine
This moves to a simple queue with no routine processing it, to reduce
scheduler pressure.
This splits latency in half!
benchmark old ns/op new ns/op delta
BenchmarkThroughput-16 2394 2364 -1.25%
BenchmarkLatency-16 259652 120810 -53.47%
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 27 Jan 2021 14:56:49 +0000 (15:56 +0100)]
tun: use %w for errors on linux
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Wed, 27 Jan 2021 14:53:21 +0000 (15:53 +0100)]
device: avoid deadlock when changing private key and removing self peers
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 26 Jan 2021 22:44:37 +0000 (23:44 +0100)]
device: use linked list for per-peer allowed-ip traversal
This makes the IpcGet method much faster.
We also refactor the traversal API to use a callback so that we don't
need to allocate at all. Avoiding allocations we do self-masking on
insertion, which in turn means that split intermediate nodes require a
copy of the bits.
benchmark old ns/op new ns/op delta
BenchmarkUAPIGet-16 3243 2659 -18.01%
benchmark old allocs new allocs delta
BenchmarkUAPIGet-16 35 30 -14.29%
benchmark old bytes new bytes delta
BenchmarkUAPIGet-16 1218 737 -39.49%
This benchmark is good, though it's only for a pair of peers, each with
only one allowedips. As this grows, the delta expands considerably.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Tue, 26 Jan 2021 22:05:48 +0000 (23:05 +0100)]
device: combine debug and info log levels into 'verbose'
There are very few cases, if any, in which a user only wants one of
these levels, so combine it into a single level.
While we're at it, reduce indirection on the loggers by using an empty
function rather than a nil function pointer. It's not like we have
retpolines anyway, and we were always calling through a function with a
branch prior, so this seems like a net gain.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Josh Bleecher Snyder [Fri, 22 Jan 2021 22:11:17 +0000 (14:11 -0800)]
device: change logging interface to use functions
This commit overhauls wireguard-go's logging.
The primary, motivating change is to use a function instead
of a *log.Logger as the basic unit of logging.
Using functions provides a lot more flexibility for
people to bring their own logging system.
It also introduces logging helper methods on Device.
These reduce line noise at the call site.
They also allow for log functions to be nil;
when nil, instead of generating a log line and throwing it away,
we don't bother generating it at all.
This spares allocation and pointless work.
This is a breaking change, although the fix required
of clients is fairly straightforward.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Tue, 26 Jan 2021 18:35:25 +0000 (10:35 -0800)]
device: fix shadowing of err in IpcHandle
The declaration of err in
nextByte, err := buffered.ReadByte
shadows the declaration of err in
op, err := buffered.ReadString('\n')
above. As a result, the assignments to err in
err = ipcErrorf(ipc.IpcErrorInvalid, "trailing character in UAPI get: %c", nextByte)
and in
err = device.IpcGetOperation(buffered.Writer)
do not modify the correct err variable.
Found by staticcheck.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Tue, 26 Jan 2021 16:40:15 +0000 (08:40 -0800)]
device: remove extra error arg
Caught by go vet.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Brad Fitzpatrick [Tue, 26 Jan 2021 19:11:58 +0000 (11:11 -0800)]
device: reduce allocs in Device.IpcGetOperation
Plenty more to go, but a start:
name old time/op new time/op delta
UAPIGet-4 6.37µs ± 2% 5.56µs ± 1% -12.70% (p=0.000 n=8+8)
name old alloc/op new alloc/op delta
UAPIGet-4 1.98kB ± 0% 1.22kB ± 0% -38.71% (p=0.000 n=10+10)
name old allocs/op new allocs/op delta
UAPIGet-4 42.0 ± 0% 35.0 ± 0% -16.67% (p=0.000 n=10+10)
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Josh Bleecher Snyder [Tue, 26 Jan 2021 19:39:48 +0000 (11:39 -0800)]
device: add benchmark for UAPI Device.IpcGetOperation
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Brad Fitzpatrick [Tue, 26 Jan 2021 02:52:42 +0000 (18:52 -0800)]
conn: fix interface parameter name in Bind interface docs
Signed-off-by: Brad Fitzpatrick <bradfitz@tailscale.com>
Jason A. Donenfeld [Mon, 25 Jan 2021 18:00:43 +0000 (19:00 +0100)]
device: allow pipelining UAPI requests
The original spec ends with \n\n especially for this reason.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Jason A. Donenfeld [Mon, 25 Jan 2021 18:08:41 +0000 (19:08 +0100)]
ipc: add missing Windows errno
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Josh Bleecher Snyder [Mon, 25 Jan 2021 17:35:35 +0000 (09:35 -0800)]
device: serialize access to IpcSetOperation
Interleaves IpcSetOperations would spell trouble.
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>
Josh Bleecher Snyder [Mon, 25 Jan 2021 17:32:09 +0000 (09:32 -0800)]
device: simplify handling of IPC set endpoint
Signed-off-by: Josh Bleecher Snyder <josh@tailscale.com>