Signed-off-by: Josh Bleecher Snyder <[email protected]>

Caught by 'go vet'.

Signed-off-by: Josh Bleecher Snyder <[email protected]>

Signed-off-by: Josh Bleecher Snyder <[email protected]>

Block instead. Backpressure here is fine, probably preferable.
This reduces code complexity.

Si...

This appears to have been an oversight.

Signed-off-by: Josh Bleecher Snyder <[email protected]>

This is similar to commit e1fa1cc5560020e67d33aa7e74674853671cf0a0,
but for the decryption chann...

This allows people to initiate connections over WireGuard without any
underlying operating syste...

Some users report seeing lines like:

> Routine: receive incoming IPv4 - stopped

Popping up une...

If Close is called after ReceiveIPvX, then ReceiveIPvX will block on an
invalid or potentially r...

It's possible for RoutineSequentialReceiver to try to lock an elem after
RoutineDecryption has e...

Persnickety consumers can now do:

func init() {
tun.WintunPool, _ = wintun.MakePool...

This is particularly problematic on mobile,
where there is a fixed number of elements.
If most o...

Picking two free ports to use for a test is difficult.
The free port we selected might no longer...

The new test introduced in this commit used to deadlock about 1% of the time.

I believe that th...

One of the first rules of WaitGroups is that you call wg.Add
outside of a goroutine, not inside ...

Signed-off-by: Josh Bleecher Snyder <[email protected]>

When running many concurrent test processing using
https://godoc.org/golang.org/x/tools/cmd/stre...

In each case, the starting waitgroup did nothing but ensure
that the goroutine has launched.

No...

Switch from *testing.T to testing.TB.

Signed-off-by: Josh Bleecher Snyder <[email protected]>

Co-authored-by: David Anderson <[email protected]>
Signed-off-by: Josh Bleecher Snyder <jo...

People are actually hitting this condition, so make it uniform. Also,
change a printf into a pri...

Signed-off-by: Brad Fitzpatrick <[email protected]>

This prevents linking to wintun.dll until it's actually needed, which
should improve startup tim...

Any io.Reader will do, and there are no performance concerns here.
This is technically backwards...

This keeps the output minimal and focused on the benchmark results.

Signed-off-by: Josh Bleeche...

This is a similar treatment to the handling of the encryption
channel found a few commits ago: U...

Make the code slightly more idiomatic. No functional changes.

Signed-off-by: Josh Bleecher Snyd...

This was shifted by 2 bytes when making persistent keepalive into a u32.
Fix it by placing it af...

SO_REUSEADDR does not make sense for unicast UDP sockets.

Signed-off-by: Jason A. Donenfeld <Ja...

Found by the race detector and existing tests.

To avoid introducing a lock into this hot path,
...

This makes it easier to work with configs in tests.
It'll see heavier use over upcoming commits;...

Since we already have it packed into a uint64
in a known byte order, write it back out again
the...

Signed-off-by: Jason A. Donenfeld <[email protected]>

Since _32 and _64 aren't valid goarchs, they don't match _GOOS_GOARCH,
and so the existing tags ...

It doesn't really matter, because it is only used in tests,
but it does remove some noise from p...

This doesn't cause any practical problems as it is,
but vet (rightly) flags this code as copying...

Signed-off-by: Josh Bleecher Snyder <[email protected]>

Minor code cleanup; no functional changes.

Signed-off-by: Josh Bleecher Snyder <josh@tailscale....

Signed-off-by: Jason A. Donenfeld <[email protected]>

Which really shouldn't happen. But it is a useful information for
troubleshooting.

Signed-off-b...

Signed-off-by: Simon Rozman <[email protected]>

Signed-off-by: Jason A. Donenfeld <[email protected]>

fix panic: send on closed channel when remove peer

Signed-off-by: Haichao Liu <liuhaichao@byted...

We let people loading this from resources opt in via:

go build -tags load_wintun_from_rsrc
...

Signed-off-by: Jason A. Donenfeld <[email protected]>

Signed-off-by: Jason A. Donenfeld <[email protected]>

Signed-off-by: Simon Rozman <[email protected]>
Signed-off-by: Jason A. Donenfeld <[email protected]>

Signed-off-by: Simon Rozman <[email protected]>
Signed-off-by: Jason A. Donenfeld <[email protected]>

Rather than having every application using Wintun driver reinvent the
wheel, the Wintun device/a...

Signed-off-by: Jason A. Donenfeld <[email protected]>

Direct syscalls using unix.Syscall(unix.SYS_*, ...) are discouraged on
macOS and might not be su...

Direct syscalls using unix.Syscall(unix.SYS_*, ...) are discouraged on
macOS and might not be su...

Direct syscalls using unix.Syscall(unix.SYS_*, ...) are discouraged on
macOS and might not be su...

This adds the fixes for golang/go#41868 which are needed to build
wireguard without direct sysca...

Direct syscalls using unix.Syscall(unix.SYS_*, ...) are discouraged on
macOS and might not be su...

Signed-off-by: Brad Fitzpatrick <[email protected]>
[Jason: ran go mod tidy.]
Signed-off-by...

Signed-off-by: Frank Werner <[email protected]>
Signed-off-by: Jason A. Donenfeld <[email protected]>

Signed-off-by: Riobard Zhan <[email protected]>
Signed-off-by: Jason A. Donenfeld <[email protected]>

Signed-off-by: Riobard Zhan <[email protected]>
Signed-off-by: Jason A. Donenfeld <[email protected]>

Signed-off-by: Jason A. Donenfeld <[email protected]>

Bits / Bytes-per-Word misses the step of also dividing by Bits-per-Byte,
which we need in order ...

Signed-off-by: Sina Siadat <[email protected]>
Signed-off-by: Jason A. Donenfeld <[email protected]>

Both wireguard-windows and wireguard-android access Bind
directly for these methods now.

Signed...

Signed-off-by: Brad Fitzpatrick <[email protected]>

Use the RTMGRP_IPV4_ROUTE const from x/sys/unix instead of using the
corresponding RTNLGRP_IPV4_...

Peers are currently removed after Device's goroutines are signaled to stop,
but without waiting ...

Signed-off-by: David Crawshaw <[email protected]>

Signed-off-by: David Crawshaw <[email protected]>

Signed-off-by: David Crawshaw <[email protected]>

Signed-off-by: Jason A. Donenfeld <[email protected]>

Another thing never tested ever.

Signed-off-by: Jason A. Donenfeld <[email protected]>

Signed-off-by: Jason A. Donenfeld <[email protected]>

This was evidently never tested before committing.

Signed-off-by: Jason A. Donenfeld <Jason@zx2...

Signed-off-by: Jason A. Donenfeld <[email protected]>

Reported-by: Jayakumar S <[email protected]>
Signed-off-by: Jason A. Donenfeld <Jason@zx2c...

In the presence of preemption, the current test may fail transiently.
This uses static test data...

Signed-off-by: Jason A. Donenfeld <[email protected]>

Due to the use of the new errors module, we now require at least 1.13
instead of 1.12.

Signed-o...

Signed-off-by: Jason A. Donenfeld <[email protected]>

Cleans up and splits out UAPIOpen to its own file.

Signed-off-by: David Crawshaw <crawshaw@tail...

os.MkdirAll never returns an os.IsExist error.

Signed-off-by: David Crawshaw <crawshaw@tailscal...

Go's GC semantics might not always guarantee the safety of this, and the
race detector gets upse...

Useful in testing when bad network stacks repeat or
batch large numbers of packets.

Signed-off-...

c85e4a410f27986a2967a49c0155633c716bf3ca introduced preliminary HWID
checking to speed up Wintun...

It was just returning "no such file or directory" (the String of the
syscall.Errno returned by C...

Signed-off-by: Simon Rozman <[email protected]>

This makes uapi.go's public API conform to Go style in terms
of error types.

Signed-off-by: Dav...

This code is useful to other packages writing tests.

Signed-off-by: David Crawshaw <crawshaw@ta...

And unexport handshake constants.

Signed-off-by: David Crawshaw <[email protected]>

The sticky socket code stays in the device package for now,
as it reaches deeply into the peer l...

Signed-off-by: Brad Fitzpatrick <[email protected]>

This works around a startup race condition when competing with
HackListener, which is trying to ...

Signed-off-by: Brad Fitzpatrick <[email protected]>

Signed-off-by: Avery Pennarun <[email protected]>

Adds a test that will fail consistently on 32-bit platforms if the
struct ever changes again to ...

This lets us include the package on those platforms in a
followup commit where we split out a co...

The existing test would occasionally flake out with:

--- FAIL: TestRatelimiter (0.12s)
ra...