aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* Minor main.go signal fixesFilippo Valsorda2018-05-211-2/+3
| | | | | | | | * Buffer the signal channel as it's non-blocking on the sender side * Notify on SIGTERM instead of the uncatchable SIGKILL License: MIT Signed-off-by: Filippo Valsorda <valsorda@google.com>
* Fix Sscanf use in tun_darwinFilippo Valsorda2018-05-211-2/+2
| | | | | License: MIT Signed-off-by: Filippo Valsorda <valsorda@google.com>
* Make successful tests silentFilippo Valsorda2018-05-212-6/+8
| | | | | License: MIT Signed-off-by: Filippo Valsorda <valsorda@google.com>
* Properly close DummyTUN to avoid deadlock in TestNoiseHandshakeFilippo Valsorda2018-05-213-1/+12
| | | | | License: MIT Signed-off-by: Filippo Valsorda <valsorda@google.com>
* User cookie is closer to fwmark than setfibJason A. Donenfeld2018-05-211-2/+2
|
* Remove broken windows cruftJason A. Donenfeld2018-05-212-533/+0
|
* Rework freebsd supportJason A. Donenfeld2018-05-2110-353/+182
|
* Add FreeBSD supportBrady OBrien2018-05-218-20/+792
| | | | Signed-off-by: Brady OBrien <brady.obrien128@gmail.com>
* Close events channel when no status listenerJason A. Donenfeld2018-05-212-3/+10
|
* Straighten out UAPI loggingJason A. Donenfeld2018-05-211-17/+17
|
* Close hack listener before closing channelJason A. Donenfeld2018-05-211-10/+14
|
* ratelimiter: do not run GC with nothing to doJason A. Donenfeld2018-05-211-8/+23
|
* Reasonable punctuation given the spacingJason A. Donenfeld2018-05-213-22/+22
|
* Fix data races in timersJason A. Donenfeld2018-05-204-17/+17
|
* Fix race with closing event channelJason A. Donenfeld2018-05-202-3/+6
| | | | | There's still a tiny race on Linux, since the tun channel is written to from two places.
* StyleJason A. Donenfeld2018-05-202-5/+4
|
* Remove unused mtu variableJason A. Donenfeld2018-05-201-2/+0
|
* Give bind its own wait groupJason A. Donenfeld2018-05-203-4/+8
| | | | In a waitgroup, all waits must come after all adds
* Avoid deadlock when the mutex isn't required, since these are atomicsJason A. Donenfeld2018-05-201-4/+0
| | | | | Maybe this fixes the "double lock issue" in f73d2fb2d96bc3fbc8bc4cce452e3c19689de01e?
* Use proper status listener on DarwinJason A. Donenfeld2018-05-201-53/+92
|
* Reduce the hack listener to once a secondJason A. Donenfeld2018-05-201-1/+1
|
* Fix race in netlink peer correlatorJason A. Donenfeld2018-05-201-0/+9
|
* Fix race in lock pendingJason A. Donenfeld2018-05-201-4/+10
|
* Fix race in statsJason A. Donenfeld2018-05-201-2/+2
|
* Fix race in packetInNonceQueueIsAwaitingKeyJason A. Donenfeld2018-05-202-7/+7
|
* Discourage building for LinuxJason A. Donenfeld2018-05-201-0/+6
|
* Revert "Temporary work around. Please revert me"0.0.20180519Jason A. Donenfeld2018-05-191-2/+0
| | | | This reverts commit 4312a7c70a58656891ff1398f2d0d36f4ed74c8a.
* timers: no need to clear keepalive in persistent keepaliveJason A. Donenfeld2018-05-191-3/+0
| | | | | | | | | | We do this after sending the keepalive anyway. This is something of a regression, though, since before we'd cancel and then send, but now we send and then cancel, so it introduces a potential race, but hopefully that isn't too big of a deal. Kernel module commit a24b3e6e15ae1ea1291666e5da910caf43eedbaf
* timers: clear send_keepalive timer on sending handshake responseJason A. Donenfeld2018-05-192-10/+14
| | | | | | | | We reorganize this into also doing so on sending keepalives itself, which means the state machine is much more consistent, even if this was already implied. Kernel module commit 30290ef1d2581a3e6ee8ffcdb05d580cfba976be
* Listen for flush in outer selectMathias Hall-Andersen2018-05-191-8/+30
| | | | | | | | Now listen for flushNonceQueue signal in outer select during the RoutineNonce routine. This is needed to handle the edge case where the queue is flushed, but no packets are in the nonce queue. Since the signal has capacity 1 this signal will remain and potentially flush the queue at a later time, with packets meant for transmission.
* Add copyright headersMathias Hall-Andersen2018-05-1938-0/+38
|
* Avoid using v6-mapped-v4Jason A. Donenfeld2018-05-182-4/+14
|
* StyleJason A. Donenfeld2018-05-181-3/+3
|
* Fill out readmeJason A. Donenfeld2018-05-171-5/+51
|
* Finer-grained start-stop synchronizationJason A. Donenfeld2018-05-166-5/+33
|
* Fix dir permissions0.0.20180514Jason A. Donenfeld2018-05-152-2/+2
|
* Temporary work around. Please revert meJason A. Donenfeld2018-05-151-0/+2
|
* Lock timers on modificationJason A. Donenfeld2018-05-151-7/+32
|
* Avoid vendoring tooling as much as possibleJason A. Donenfeld2018-05-153-32/+24
|
* Add vendoring for HomebrewJason A. Donenfeld2018-05-154-1/+58
|
* Do not send keepalive if closedJason A. Donenfeld2018-05-151-1/+1
|
* Fix refactoring mistakesJason A. Donenfeld2018-05-152-3/+3
|
* Improve makefileJason A. Donenfeld2018-05-151-2/+10
|
* Use /dev/null as place holderJason A. Donenfeld2018-05-141-0/+5
|
* Print utun name on darwinJason A. Donenfeld2018-05-142-2/+6
|
* Mask IPs when showing from trieJason A. Donenfeld2018-05-141-16/+6
| | | | | Otherwise intermediate nodes that get collapsed to real nodes will display the wrong value.
* Ugly hack to suppress warning on backgrounded processJason A. Donenfeld2018-05-141-0/+4
|
* Better common bits functionJason A. Donenfeld2018-05-142-56/+44
|
* Fix noise testJason A. Donenfeld2018-05-141-7/+9
|
* Cancelable netlink writes and better response correlationJason A. Donenfeld2018-05-141-18/+22
|
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.