Commit message (Collapse) | Author | Files | Lines | ||
---|---|---|---|---|---|
2017-11-14 | curve25519-neon: compile in thumb mode | Jason A. Donenfeld | 3 | -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. | |||||
2017-11-14 | compat: 3.16.50 got proper rt6_get_cookie | Jason A. Donenfeld | 1 | -1/+1 | |
2017-11-14 | qemu: update kernel | Jason A. Donenfeld | 1 | -1/+1 | |
2017-11-13 | allowedips: do not write out of bounds | Jason A. Donenfeld | 1 | -2/+4 | |
2017-11-13 | selftest: allowedips: randomized test mutex update | Jason A. Donenfeld | 1 | -5/+6 | |
2017-11-12 | wg-quick: document localhost exception and v6 rule | Jason A. Donenfeld | 1 | -3/+4 | |
Reported-by: Hermann Lienstromberg <nurtic-vibe@grmml.net> | |||||
2017-11-11 | version: bump snapshot0.0.20171111 | Jason A. Donenfeld | 2 | -2/+2 | |
2017-11-11 | tools: allow for NULL keys everywhere | Jason A. Donenfeld | 5 | -29/+50 | |
2017-11-11 | curve25519: reject deriving from NULL private keys | Jason A. Donenfeld | 2 | -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. | |||||
2017-11-11 | tools: remove ioctl cruft | Jason A. Donenfeld | 1 | -4/+0 | |
2017-11-10 | allowedips: rename from routingtable | Jason A. Donenfeld | 11 | -149/+149 | |
Makes it more clear that this _not_ a routing table replacement. | |||||
2017-11-10 | wg-quick: allow for tabs in keys | Jason A. Donenfeld | 2 | -20/+20 | |
2017-11-10 | netlink: make sure we reserve space for NLMSG_DONE | Jason A. Donenfeld | 2 | -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. | |||||
2017-11-10 | compat: 4.4.0 has strange ECN function | Jason A. Donenfeld | 1 | -1/+2 | |
2017-11-10 | wg-quick: stat the correct enclosing folder of config file | Jason A. Donenfeld | 1 | -1/+2 | |
2017-11-10 | qemu: bump kernel version | Jason A. Donenfeld | 1 | -1/+1 | |
2017-11-10 | receive: hoist fpu outside of receive loop | Jason A. Donenfeld | 3 | -18/+19 | |
2017-11-10 | qemu: more debugging | Jason A. Donenfeld | 3 | -1/+15 | |
2017-11-10 | device: wait for all peers to be freed before destroying | Jason A. Donenfeld | 1 | -0/+1 | |
2017-11-10 | qemu: check for memory leaks | Jason A. Donenfeld | 3 | -6/+28 | |
2017-11-03 | netlink: plug memory leak | Jason A. Donenfeld | 1 | -1/+1 | |
2017-11-03 | device: please lockdep | Jason A. Donenfeld | 1 | -1/+1 | |
2017-11-02 | global: revert checkpatch.pl changes | Jason A. Donenfeld | 2 | -2/+2 | |
These changes were suggested by checkpatch.pl, but actually cause big problems depending on the options. Revert. | |||||
2017-11-02 | Kconfig: remove trailing whitespace | Greg Kroah-Hartman | 1 | -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> | |||||
2017-11-01 | version: bump snapshot0.0.20171101 | Jason A. Donenfeld | 2 | -2/+2 | |
2017-11-01 | compat: unbreak unloading on kernels 4.6 through 4.9 | Jason A. Donenfeld | 1 | -1/+4 | |
2017-11-01 | timers: switch to kees' new timer_list functions | Jason A. Donenfeld | 2 | -17/+22 | |
2017-11-01 | wg-quick: save all hooks on save | Jason A. Donenfeld | 1 | -5/+13 | |
2017-10-31 | version: bump snapshot0.0.20171031 | Jason A. Donenfeld | 2 | -2/+2 | |
2017-10-31 | wg-quick: fsync the temporary file before renaming | Jason A. Donenfeld | 1 | -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> | |||||
2017-10-31 | wg-quick: allow for saving existing interface | Jason A. Donenfeld | 2 | -2/+14 | |
2017-10-31 | contrib: add reresolve-dns | Jason A. Donenfeld | 2 | -0/+48 | |
2017-10-31 | tools: correct type for CTRL_ATTR_FAMILY_ID | Jason A. Donenfeld | 1 | -4/+4 | |
Suggested-by: Jörg Thalheim <joerg@thalheim.io> | |||||
2017-10-31 | wg-quick: allow for the hatchet, but not by default | Jason A. Donenfeld | 4 | -0/+55 | |
2017-10-31 | global: use fewer BUG_ONs | Jason A. Donenfeld | 2 | -3/+5 | |
Suggested-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> | |||||
2017-10-31 | timers: guard entire setting in block | Jason A. Donenfeld | 1 | -4/+6 | |
2017-10-31 | curve25519: only enable int128 if compiler support is sound | Jason A. Donenfeld | 1 | -1/+1 | |
2017-10-31 | device: expand scope of destruct lock | Jason A. Donenfeld | 1 | -2/+2 | |
2017-10-31 | global: get rid of useless forward declarations | Jason A. Donenfeld | 3 | -11/+0 | |
2017-10-31 | device: only take reference if netns is different | Jason A. Donenfeld | 3 | -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. | |||||
2017-10-31 | wg-quick: remember to rewind DNS settings on failure | Jason A. Donenfeld | 1 | -3/+8 | |
2017-10-31 | wg-quick: allow specifiying multiple hooks | Jason A. Donenfeld | 2 | -18/+21 | |
2017-10-31 | qemu: test using four cores | Jason A. Donenfeld | 1 | -1/+1 | |
2017-10-31 | global: style nits | Jason A. Donenfeld | 27 | -274/+470 | |
2017-10-31 | qemu: work around ccache bugs | Jason A. Donenfeld | 1 | -1/+1 | |
2017-10-31 | global: infuriating kernel iterator style | Jason A. Donenfeld | 13 | -48/+48 | |
One types: for (i = 0 ... So one should also type: for_each_obj (obj ... But the upstream kernel style guidelines are insane, and so we must instead do: for_each_obj(obj ... Ugly, but one must choose his battles wisely. | |||||
2017-10-31 | peer: store total number of peers instead of iterating | Jason A. Donenfeld | 3 | -15/+4 | |
This is faster, since it means adding a new peer is O(1) instead of O(n). It's also safe to do because we're holding the device_update_lock on both the ++ and the --. | |||||
2017-10-31 | peer: get rid of peer_for_each magic | Jason A. Donenfeld | 3 | -41/+17 | |
Since the peer list is protected by the device_update_lock, and since items are removed from the peer list before putting their final reference, we don't actually need to take a reference when iterating. This allows us to simplify the macro considerably. Suggested-by: Johannes Berg <johannes@sipsolutions.net> | |||||
2017-10-31 | compat: be sure to include header before testing | Jason A. Donenfeld | 1 | -0/+1 | |
Reported-by: Aaron M. D. Jones <aaronmdjones@gmail.com> | |||||
2017-10-31 | qemu: allow for cross compilation | Jason A. Donenfeld | 14 | -98/+249 | |