summaryrefslogtreecommitdiffhomepage
path: root/src
AgeCommit message (Collapse)Author
2016-12-23cookies: use xchacha20poly1305 instead of chacha20poly1305Jason A. Donenfeld
This allows us to precompute the blake2s calls and save cycles, since hchacha is fast. Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-23routing-table: simplify and mask reparented rootJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-23config: allow removing multiple peers at onceJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-18version: bump snapshotJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-16siphash: preserve endian-ness for quick helperJason A. Donenfeld
This fixes errors on big endian machines. Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-16ratelimiter: remember newline in errorJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-16version: bump snapshotJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-16hashtables: use counter and int to ensure forward progressJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-16messages: increase header by 3 bytes for alignmentJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-16siphash: update against upstream submissionJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-16ratelimiter: drop family from action for 4.10Jason A. Donenfeld
This was never actually used, and now that member of the struct has been removed all together from the kernel. Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-16tests: avoid non-strict writes via printfJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-16noise: update commentsJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-16types: enforce consistencyJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-16config: cleanupsJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-13peer: don't use sockaddr_storage to reduce memory usageJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-12hashtables: ensure we get 64-bits of randomnessJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-12messages: remove unused constantsJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-12siphash: useless riceJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11crypto: use kernel's bitops functionsJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11blake2s: move self tests to correct directoryJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11global: move to consistent use of uN instead of uintN_t for kernel codeJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11receive: simplify ip header checking logicJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11device: ensure icmp skb length check is done for v6Jason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11device: allocate tstats in newlinkJason A. Donenfeld
Every place else does this in ndo_init, but this seems completely redundant, so to simplify things, I'm moving it into newlink. Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11device: clean up xmit error pathJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11main: consistent linesJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11device: simplify device_init, since it only returns -EEXISTJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11device: dellink is already implictly unregister_netdevice_queueJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11device: disable ipv6 auto address generationJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-11device: rc -> retJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09version: bump snapshotJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09cookie: kill redundant forward declarationJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09compat: build dep errors belong here, since it's out of tree specificJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09socket: clear src address when retrying handshakeJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09device: traditional if is cleaner than switch for this smallJason A. Donenfeld
This also generates identical assembly. Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09device: make suspend code conditional on CONFIG_PM_SLEEPJason A. Donenfeld
This isn't actually neccessary, since there are no-op stubs for these functions, but the MIPS people don't want any unneccessary bloat. Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09main: cleaner error teardownJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09device: cleaner error teardownJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09timers: add random jitter to handshake retryJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09device: clear all peer ephemeral keys on sleepJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09data: reset tc when resetting skbJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09qemu: bump kernel versionJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-09build system: add dkms installationJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-04tests: directly kill nmapJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-12-04tests: make sure ncat gets killedJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-11-29hashtable: use random number each timeJason A. Donenfeld
Otherwise timing information might leak information about prior index entries. We also switch back to an explicit uint64_t because siphash needs something at least that size. (This partially reverts 1550e9ba597946c88e3e7e3e8dcf33c13dd76e5b. Willy's suggestion was wrong.) Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-11-29version: bump snapshotJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-11-29send: send packet initiation only after requeuing to prevent raceJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
2016-11-29tests: be sure we get all messagesJason A. Donenfeld
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>