* send: switch handshake stamp to an atomic
Rather than abusing the handshake lock, we're much better off just using
a boring atomic64 for this. It's simpler and performs better. Also, while
we're at it, we set the handshake stamp both before and after the
calculations, in case the calculations block for a really long time waiting
for the RNG to initialize.
* compat: better atomic acquire/release backport
This should fix compilation and correctness on several platforms.
* crypto: move simd context to specific type
This was a suggestion from Andy Lutomirski on LKML.
* chacha20poly1305: selftest: use arrays for test vectors
We no longer have lines so long that they're rejected by SMTP servers.
* qemu: add easy git harness
This makes it a bit easier to use our qemu harness for testing our mainline
integration tree.
* curve25519-x86_64: avoid use of r12
This causes problems with RAP and KERNEXEC for PaX, as r12 is a
reserved register.
* chacha20: use memmove in case buffers overlap
A small correctness fix that we never actually hit in WireGuard but is
important especially for moving this into a general purpose library.
* curve25519-hacl64: simplify u64_eq_mask
* curve25519-hacl64: correct u64_gte_mask
Two bitmath fixes from Samuel, which come complete with a z3 script proving
their correctness.
* timers: include header in right file
This fixes compilation in some environments.
* netlink: don't start over iteration on multipart non-first allowedips
Matt Layher found a bug where a netlink dump of peers would never terminate in
some circumstances, causing wg(8) to keep trying forever. We now have a fix as
well as a unit test to mitigate this, and we'll be looking to create a fuzzer
out of Matt's nice library.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Unauthenticated EAPOL-Key decryption in wpa_supplicant
Published: August 8, 2018
Identifiers:
- CVE-2018-14526
Latest version available from: https://w1.fi/security/2018-1/
Vulnerability
A vulnerability was found in how wpa_supplicant processes EAPOL-Key
frames. It is possible for an attacker to modify the frame in a way that
makes wpa_supplicant decrypt the Key Data field without requiring a
valid MIC value in the frame, i.e., without the frame being
authenticated. This has a potential issue in the case where WPA2/RSN
style of EAPOL-Key construction is used with TKIP negotiated as the
pairwise cipher. It should be noted that WPA2 is not supposed to be used
with TKIP as the pairwise cipher. Instead, CCMP is expected to be used
and with that pairwise cipher, this vulnerability is not applicable in
practice.
When TKIP is negotiated as the pairwise cipher, the EAPOL-Key Key Data
field is encrypted using RC4. This vulnerability allows unauthenticated
EAPOL-Key frames to be processed and due to the RC4 design, this makes
it possible for an attacker to modify the plaintext version of the Key
Data field with bitwise XOR operations without knowing the contents.
This can be used to cause a denial of service attack by modifying
GTK/IGTK on the station (without the attacker learning any of the keys)
which would prevent the station from accepting received group-addressed
frames. Furthermore, this might be abused by making wpa_supplicant act
as a decryption oracle to try to recover some of the Key Data payload
(GTK/IGTK) to get knowledge of the group encryption keys.
Full recovery of the group encryption keys requires multiple attempts
(128 connection attempts per octet) and each attempt results in
disconnection due to a failure to complete the 4-way handshake. These
failures can result in the AP/network getting disabled temporarily or
even permanently (requiring user action to re-enable) which may make it
impractical to perform the attack to recover the keys before the AP has
already changes the group keys. By default, wpa_supplicant is enforcing
at minimum a ten second wait time between each failed connection
attempt, i.e., over 20 minutes waiting to recover each octet while
hostapd AP implementation uses 10 minute default for GTK rekeying when
using TKIP. With such timing behavior, practical attack would need large
number of impacted stations to be trying to connect to the same AP to be
able to recover sufficient information from the GTK to be able to
determine the key before it gets changed.
Vulnerable versions/configurations
All wpa_supplicant versions.
Acknowledgments
Thanks to Mathy Vanhoef of the imec-DistriNet research group of KU
Leuven for discovering and reporting this issue.
Possible mitigation steps
- Remove TKIP as an allowed pairwise cipher in RSN/WPA2 networks. This
can be done also on the AP side.
- Merge the following commits to wpa_supplicant and rebuild:
WPA: Ignore unauthenticated encrypted EAPOL-Key data
This patch is available from https://w1.fi/security/2018-1/
- Update to wpa_supplicant v2.7 or newer, once available
Signed-off-by: John Crispin <john@phrozen.org>
Apply IPv6/ND configuration before proto_send_update so that all config info
is available when netifd is handling the notify_proto ubus call.
In particular this fixes an issue when netifd is updating the downstream IPv6 mtu
as netifd was still using the not yet updated upstream IPv6 mtu to set the
downstream IPv6 mtu
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
Commit 4d961538f6 added libutil to the iproute2 InstallDev section
but lead to compile issues with packages picking up the wrong libutil
since libutil is quite a generic name ...
Further libutil is rather meant for internal usage in iproute2 than a
public API; therefore let's remove it from the InstallDev section together
with ll_map.h
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
In iproute2 v4.17 ll_map has been moved from the libnetlink to the libutil
library; add libutil as well to the staging dir in order to keep support
for ll_map
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
Changelog taken from the version announcement
> == Changes ==
>
> * chacha20poly1305: selftest: split up test vector constants
>
> The test vectors are encoded as long strings -- really long strings -- and
> apparently RFC821 doesn't like lines longer than 998.
> https://cr.yp.to/smtp/message.html
>
> * queueing: keep reference to peer after setting atomic state bit
>
> This fixes a regression introduced when preparing the LKML submission.
>
> * allowedips: prevent double read in kref
> * allowedips: avoid window of disappeared peer
> * hashtables: document immediate zeroing semantics
> * peer: ensure resources are freed when creation fails
> * queueing: document double-adding and reference conditions
> * queueing: ensure strictly ordered loads and stores
> * cookie: returned keypair might disappear if rcu lock not held
> * noise: free peer references on failure
> * peer: ensure destruction doesn't race
>
> Various fixes, as well as lots of code comment documentation, for a
> small variety of the less obvious aspects of object lifecycles,
> focused on correctness.
>
> * allowedips: free root inside of RCU callback
> * allowedips: use different macro names so as to avoid confusion
>
> These incorporate two suggestions from LKML.
>
> This snapshot contains commits from: Jason A. Donenfeld and Jann Horn.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
This patch makes sch_cake's gso/gro splitting configurable
from userspace.
To disable breaking apart superpackets in sch_cake:
tc qdisc replace dev whatever root cake no-split-gso
to enable:
tc qdisc replace dev whatever root cake split-gso
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: Dave Taht <dave.taht@gmail.com>
[pulled from netdev list - no API/ABI change]
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Fixes the annoying 'feature' were TTL was set to "1" by default ..
Users had to specify -T manually to test outside the own network.
2.0.12 change set (as of June 25th 2018)
o Change the unicast TTL default value from 1 to the system default (to be compatable with previous versions.) Multicast still defaults to 1.
o adpative formatting bug fix: crash occurs when values exceed 1 Tera. Add support for Tera and Peta and eliminate the potential crash condition
o configure default compile to include isochronous support (use configure --disable-isochronous to remove support)
o replace 2.0.11's --vary-load option with a more general -b option to include <mean>,<stdev>, e.g. -b 100m,40m, which will pull from a log normal distribution every 0.1 seconds
o fixes for windows cross compile (using mingw32)
o compile flags of -fPIE for android
o configure --enable-checkprograms to compile ancillary binaries used to test things such as delay, isoch, pdf generation
o compile tests when trying to use 64b seq numbers on a 32b platform
o Fix GCC ver 8 warnings
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
a514139 build: compile with -ffunction-sections, -fdata-sections and LTO
3c30b17 wl: only invoke nvram executable if it exists
65b8333 Revert "build: compile with -ffunction-sections, -fdata-sections and LTO"
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
75ee790 interface-ip: fix eui64 ifaceid generation (FS#1668)
ca97097 netifd: make sure the vlan ifname fits into the buffer
b8c1bca iprule: remove bogus assert calls
a2f952d iprule: fix broken in_dev/out_dev checks
263631a vlan: use alloca to get rid of IFNAMSIZE in vlan_dev_set_name()
291ccbb ubus: display correct prefix size for IPv6 prefix address
908a9f4 CMakeLists.txt: add -Wimplicit-fallthrough to the compiler flags
b06b011 proto-shell.c: add a explicit "fall through" comment to make the compiler happy
60293a7 replace fall throughs in switch/cases where possible with simple code changes
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
When libcap-ng is detected during build, support for it is enabled. This
will cause a build failure due to a missing dependency. Explicitly
disable libcap-ng support to avoid this.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
The call "get_features" allows to gather hostapd config options
via ubus. As first infos we add the ht and vht support.
Although nl80211 supports to gather informations about
ht and vht capabilities, the hostapd configuration can disable
vht and ht. However, it is possible that the iw output is not
representing the actual hostapd configuration.
Signed-off-by: Nick Hainke <vincent@systemli.org>
Upstream renamed openssl-1.0.cnf to openssl-easyrsa.cnf.
However, pkg kept using openssl-1.0.cnf.
Upstream easyrsa searchs for vars, openssl-*, x509-types in the
same directory as easyrsa script. This was patched to revert
back to static /etc/easy-rsa/ directory (as does OpenSUSE).
EASYRSA_PKI still depends on $PWD.
Move easyrsa from /usr/sbin to /usr/bin as root is not needed.
Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
It is insecure to let this type of packets inside
They can e.g. open ports on some other routers with UPnP, etc
Signed-off-by: Dmitry Tunin <hanipouspilot@gmail.com>
[0-3](none, minimal[default], more, maximum)
It is not 100% backward compatible, because now 0 disables logging
Signed-off-by: Dmitry Tunin <hanipouspilot@gmail.com>
Refresh patches
Upstream commits since last bump:
3b6eb19 Log DNSSEC trust anchors at startup.
f3e5787 Trivial comment change.
c851c69 Log failure to confirm an address in DHCPv6.
a3bd7e7 Fix missing fatal errors when parsing some command-line/config options.
ab5ceaf Document the --help option in the french manual
1f2f69d Fix recurrent minor spelling mistake in french manual
f361b39 Fix some mistakes in french translation of the manual
eb1fe15 When replacing cache entries, preserve CNAMES which target them.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
The sierra_net driver is using proto_directip_setup for setup. So use
proto_directip_teardown for teardown.
Signed-off-by: Masashi Honma <masashi.honma@gmail.com>
Disabled libpsl to fix build issue reported by buildbots
Package libcurl is missing dependencies for the following libraries:
libpsl.so.5
Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
This watchdog script tries to re-resolve hostnames for inactive WireGuard peers.
Use it for peers with a frequently changing dynamic IP.
persistent_keepalive must be set, recommended value is 25 seconds.
Run this script from cron every minute:
echo '* * * * * /usr/bin/wireguard_watchdog' >> /etc/crontabs/root
Signed-off-by: Aleksandr V. Piskunov <aleksandr.v.piskunov@gmail.com>
[bump the package release]
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
80b41cd version: bump snapshot
fe5f0f6 recieve: disable NAPI busy polling
e863f40 device: destroy workqueue before freeing queue
81a2e7e wg-quick: allow link local default gateway
95951af receive: use gro call instead of plain call
d9501f1 receive: account for zero or negative budget
e80799b tools: only error on wg show if all interfaces failk
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
[Added commit log to commit description]
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
iproute2's tc was updated to support the recently upstreamed cake qdisc.
Backport this canonical support from upstream into iproute2 v4.17
There is no kernel kmod/userspace tc ABI change in this release from the
previous package bump, so everyone can breath a sigh of relief.
This is largely a code style change, the exception to prove the rule:
option 'autorate_ingress' has been changed to 'autorate-ingress' to fit
in with upstream option naming expectations.
No openwrt package (e.g. sqm-scripts) has knowledge of
'autorate_ingress' thus only users who made their own scripts or used
it within the 'dangerous configuration' options of sqm-scripts will be
affected.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Add each variant to the matching PROVIDERS variables after evaluating
the respective hostapd*, wpad* and wpa* variant.
Each package providing the same feature will automatically conflict with
all prior packages providing the same feature.
This way we can handle the conflicts automatically without introducing
recursive dependencies.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Move common variables and/or values to the package (variant) default.
Add additional values in variant packages if necessary. Remove further
duplicates by introducing new templates.
Remove the ANY_[HOSTAPD|SUPPLICANT_PROVIDERS]_PROVIDERS. The are the
same as the variables without the any prefix. No need to maintain both
variables.
Signed-off-by: Mathias Kresin <dev@kresin.me>
procd needs processes to stay in foreground to remain under its gaze and
control. Failure to do so means service stop commands fail to actually
stop the process (procd doesn't think it's running 'cos the process has
exited already as part of its forking routing)
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
As dnsmasq is started earlier than netifd usage of network.sh functions
at boottime will fail; therefore don't call at boottime the functions
which construct the dhcp pool/relay info.
As interface triggers are installed the dhcp pool/relay info will be
constructed when the interface gets reported as up by netifd.
At the same time also register interface triggers based on DHCP relay
config.
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
The pptp.so plugin needs to be built with -fPIC as well in order to be
linkable again.
Fixes 888a15ff83 ("ppp: add missing -fPIC to rp-pppoe.so CFLAGS")
Fixes e7397eef69 ("ppp: compile with LTO enabled")
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Increase the termination timeout to 15s to let OpenVPN properly tear down
its connections, especially when weak links or complex down scripts are
involved.
Fixes FS#859.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Override the default shutdown action (stop) and close all processes
of dropbear
Since commit 498fe85, the stop action only closes the process
that's listening for new connections, maintaining the ones with
existing clients.
This poses a problem when restarting or shutting-down a device,
because the connections with existing SSH clients, like OpenSSH,
are not properly closed, causing them to hang.
This situation can be avoided by closing all dropbear processes when
shutting-down the system, which closes properly the connections with
current clients.
Signed-off-by: Christian Schoenebeck <christian.schoenebeck@gmail.com>
[Luis: Rework commit message]
Signed-off-by: Luis Araneda <luaraneda@gmail.com>
Fix broken DHCPv6 servers which provide the server unicast option but
do not reply on DHCPv6 renew messages directed to the IPv6 address
contained in the server unicast option whihc results in broken IPv6
connectivity.
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
