aboutsummaryrefslogtreecommitdiffstatshomepage
path: root/src (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* chacha20poly1305: add more test vectors, some of which are weirdJason A. Donenfeld2017-11-221-3/+244
|
* compat: new kernels have netlink fixesJason A. Donenfeld2017-11-222-13/+10
|
* compat: stable finally backported fixJason A. Donenfeld2017-11-221-1/+1
|
* qemu: use unprefixed strip when not cross-compilingJason A. Donenfeld2017-11-171-1/+2
| | | | | | | Fedora/Redhat doesn't ship with a prefixed strip, and we don't need to use it anyway when we're not cross compiling, so don't. Reported-by: Samuel Neves <sneves@dei.uc.pt>
* tools: tighten up strtoul parsingJason A. Donenfeld2017-11-172-36/+62
| | | | Reported-by: Cedric Buxin <cedric.buxin@izri.org>
* device: uninitialize socket first in destructionJason A. Donenfeld2017-11-171-2/+2
|
* socket: only free socket after successful creation of newJason A. Donenfeld2017-11-174-47/+35
| | | | | | | | | | | | | | When an interface is down, the socket port can change freely. A socket will be allocated when the interface comes up, and if a socket can't be allocated, the interface doesn't come up. However, a socket port can change while the interface is up. In this case, if a new socket with a new port cannot be allocated, it's important to keep the interface in a consistent state. The choices are either to bring down the interface or to preserve the old socket. This patch implements the latter. Reported-by: Marc-Antoine Perennou <keruspe@exherbo.org>
* compat: fix compilation with PaXJason A. Donenfeld2017-11-141-0/+12
| | | | | | Despite not having access to the latest grsecurity source, I still do very much want to keep supporting it, so this patch is my best guess at what will work.
* curve25519-neon: compile in thumb modeJason A. Donenfeld2017-11-143-10/+7
| | | | | In thumb mode, it's not possible to use sp as an operand of and, so we have to muck around with r3 as a scratch register.
* compat: 3.16.50 got proper rt6_get_cookieJason A. Donenfeld2017-11-141-1/+1
|
* qemu: update kernelJason A. Donenfeld2017-11-141-1/+1
|
* allowedips: do not write out of boundsJason A. Donenfeld2017-11-131-2/+4
|
* selftest: allowedips: randomized test mutex updateJason A. Donenfeld2017-11-131-5/+6
|
* wg-quick: document localhost exception and v6 ruleJason A. Donenfeld2017-11-121-3/+4
| | | | Reported-by: Hermann Lienstromberg <nurtic-vibe@grmml.net>
* version: bump snapshot0.0.20171111Jason A. Donenfeld2017-11-112-2/+2
|
* tools: allow for NULL keys everywhereJason A. Donenfeld2017-11-115-29/+50
|
* curve25519: reject deriving from NULL private keysJason A. Donenfeld2017-11-112-7/+16
| | | | | These aren't actually valid 25519 points pre-normalization, and doing this is required to make unsetting private keys based on all zeros.
* tools: remove ioctl cruftJason A. Donenfeld2017-11-111-4/+0
|
* allowedips: rename from routingtableJason A. Donenfeld2017-11-1011-149/+149
| | | | Makes it more clear that this _not_ a routing table replacement.
* wg-quick: allow for tabs in keysJason A. Donenfeld2017-11-101-2/+2
|
* netlink: make sure we reserve space for NLMSG_DONEJason A. Donenfeld2017-11-102-1/+25
| | | | | | | Otherwise, if messages pack really close together, we'll exceed the size of the sk_buff and return ENOBUFS. I suspect this has been the cause of a lot of weird bugs that people just worked around by increasing the receive buffer size. This actually addresses the root cause.
* compat: 4.4.0 has strange ECN functionJason A. Donenfeld2017-11-101-1/+2
|
* wg-quick: stat the correct enclosing folder of config fileJason A. Donenfeld2017-11-101-1/+2
|
* qemu: bump kernel versionJason A. Donenfeld2017-11-101-1/+1
|
* receive: hoist fpu outside of receive loopJason A. Donenfeld2017-11-103-18/+19
|
* qemu: more debuggingJason A. Donenfeld2017-11-103-1/+15
|
* device: wait for all peers to be freed before destroyingJason A. Donenfeld2017-11-101-0/+1
|
* qemu: check for memory leaksJason A. Donenfeld2017-11-103-6/+28
|
* netlink: plug memory leakJason A. Donenfeld2017-11-031-1/+1
|
* device: please lockdepJason A. Donenfeld2017-11-031-1/+1
|
* global: revert checkpatch.pl changesJason A. Donenfeld2017-11-022-2/+2
| | | | | These changes were suggested by checkpatch.pl, but actually cause big problems depending on the options. Revert.
* Kconfig: remove trailing whitespaceGreg Kroah-Hartman2017-11-021-1/+1
| | | | | | | Git complains when applying this Kconfig file as part of a patch to the kernel tree, so drop the trailing whitespace. Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* version: bump snapshot0.0.20171101Jason A. Donenfeld2017-11-012-2/+2
|
* compat: unbreak unloading on kernels 4.6 through 4.9Jason A. Donenfeld2017-11-011-1/+4
|
* timers: switch to kees' new timer_list functionsJason A. Donenfeld2017-11-012-17/+22
|
* wg-quick: save all hooks on saveJason A. Donenfeld2017-11-011-5/+13
|
* version: bump snapshot0.0.20171031Jason A. Donenfeld2017-10-312-2/+2
|
* wg-quick: fsync the temporary file before renamingJason A. Donenfeld2017-10-311-0/+1
| | | | | | | This ensures that on an unclean shutdown, we either see the old content or the new content, but not empty content. Suggested-by: Ka Ho Ng <ngkaho1234@gmail.com>
* wg-quick: allow for saving existing interfaceJason A. Donenfeld2017-10-312-2/+14
|
* tools: correct type for CTRL_ATTR_FAMILY_IDJason A. Donenfeld2017-10-311-4/+4
| | | | Suggested-by: Jörg Thalheim <joerg@thalheim.io>
* wg-quick: allow for the hatchet, but not by defaultJason A. Donenfeld2017-10-311-0/+2
|
* global: use fewer BUG_ONsJason A. Donenfeld2017-10-312-3/+5
| | | | Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
* timers: guard entire setting in blockJason A. Donenfeld2017-10-311-4/+6
|
* curve25519: only enable int128 if compiler support is soundJason A. Donenfeld2017-10-311-1/+1
|
* device: expand scope of destruct lockJason A. Donenfeld2017-10-311-2/+2
|
* global: get rid of useless forward declarationsJason A. Donenfeld2017-10-313-11/+0
|
* device: only take reference if netns is differentJason A. Donenfeld2017-10-313-9/+56
| | | | | | If we take two references, the namespace and the device are never freed in the usual manner. We should thus only take a reference to another namespace when it is a different namespace from our own.
* wg-quick: remember to rewind DNS settings on failureJason A. Donenfeld2017-10-311-3/+8
|
* wg-quick: allow specifiying multiple hooksJason A. Donenfeld2017-10-312-18/+21
|
* qemu: test using four coresJason A. Donenfeld2017-10-311-1/+1
|
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.