crypto_bignum_rand() use needless time-consuming filtering
which resulted in SAE no longer connecting within time limits.
Import fixes from hostap upstream to fix that.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
(cherry picked from commit 631c437a91)
* compat: timeconst.h is a generated artifact
Before we were trying to check for timeconst.h by looking in the kernel
source directory. This isn't quite correct on configurations in which
the object directory is separate from the kernel source directory, for
example when using O="elsewhere" as a make option when building the
kernel. The correct fix is to use $(CURDIR), which should point to
where we want.
* compat: use bash instead of bc for HZ-->USEC calculation
This should make packaging somewhat easier, as bash is generally already
available (at least for dkms), whereas bc isn't provided by distros by
default in their build meta packages.
* socket: remove errant restriction on looping to self
It's already possible to create two different interfaces and loop
packets between them. This has always been possible with tunnels in the
kernel, and isn't specific to wireguard. Therefore, the networking stack
already needs to deal with that. At the very least, the packet winds up
exceeding the MTU and is discarded at that point. So, since this is
already something that happens, there's no need to forbid the not very
exceptional case of routing a packet back to the same interface; this
loop is no different than others, and we shouldn't special case it, but
rather rely on generic handling of loops in general. This also makes it
easier to do interesting things with wireguard such as onion routing.
At the same time, we add a selftest for this, ensuring that both onion
routing works and infinite routing loops do not crash the kernel. We
also add a test case for wireguard interfaces nesting packets and
sending traffic between each other, as well as the loop in this case
too. We make sure to send some throughput-heavy traffic for this use
case, to stress out any possible recursion issues with the locks around
workqueues.
* send: cond_resched() when processing tx ringbuffers
Users with pathological hardware reported CPU stalls on CONFIG_
PREEMPT_VOLUNTARY=y, because the ringbuffers would stay full, meaning
these workers would never terminate. That turned out not to be okay on
systems without forced preemption. This commit adds a cond_resched() to
the bottom of each loop iteration, so that these workers don't hog the
core. We don't do this on encryption/decryption because the compat
module here uses simd_relax, which already includes a call to schedule
in preempt_enable.
* selftests: initalize ipv6 members to NULL to squelch clang warning
This fixes a worthless warning from clang.
* send/receive: use explicit unlikely branch instead of implicit coalescing
Some code readibility cleanups.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
(cherry picked from commit 4f6343ffe7)
As announced on the mailing list, WireGuard will be in Linux 5.6. As a
result, the wg(8) tool, used by OpenWRT in the same manner as ip(8), is
moving to its own wireguard-tools repo. Meanwhile, the out-of-tree
kernel module for kernels 3.10 - 5.5 moved to its own wireguard-linux-
compat repo. Yesterday, releases were cut out of these repos, so this
commit bumps packages to match. Since wg(8) and the compat kernel module
are versioned and released separately, we create a wireguard-tools
Makefile to contain the source for the new tools repo. Later, when
OpenWRT moves permanently to Linux 5.6, we'll drop the original module
package, leaving only the tools. So this commit shuffles the build
definition around a bit but is basically the same idea as before.
Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
(cherry picked from commit ea980fb9c6)
Fixes: 5e8b50da15 (odhcpd : fix lan host reachibility due to identical RIO and PIO prefixes (FS#3056))
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
975dce2 client: allow keep-alive for POST requests
d062f85 file: poke ustream after starting deferred program
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Backport patch from hostapd.git master that fixes copy/paste error in
crypto_bignum_sub() in crypto_wolfssl.c.
This missing fix was discovered while testing SAE over a mesh interface.
With this fix applied and wolfssl >3.14.4 mesh+SAE works fine with
wpad-mesh-wolfssl.
Cc: Sean Parkinson <sean@wolfssl.com>
Signed-off-by: Antonio Quartulli <a@unstable.cc>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
(cherry picked from commit 4b3b8ec81c)
Without this change, wpa-cli features depend on which wpad build variant was
used to build the wpa-cli package
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Tested-by: Jérôme Benoit <jerome.benoit@piment-noir.org> [WRT1900AC v1]
[added missing package version bump]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 03e9e4ba9e)
f4d759b dhcp.c: further improve validation
Further improve input validation for CVE-2020-11752
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 9e7d11f3e2)
cdac046 dns.c: fix input validation fix
Due to a slight foobar typo, failing to de-reference a pointer, previous
fix not quite as complete as it should have been.
Improve CVE-2020-11750 fix
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 9f7c8ed078)
Fix the test for an enabled sysntp initscript in dnsmasq.init, and get
rid of "test -o" while at it.
Issue reproduced on openwrt-19.07 with the help of pool.ntp.br and an
RTC-less ath79 router. dnssec-no-timecheck would be clearly missing
from /var/etc/dnsmasq.conf.* while the router was still a few days in
the past due to non-working DNSSEC + DNS-based NTP server config.
The fix was tested with the router in the "DNSSEC broken state": it
properly started dnsmasq in dnssec-no-timecheck mode, and eventually ntp
was able to resolve the server name to an IP address, and set the system
time. DNSSEC was then enabled by SIGINT through the ntp hotplug hook,
as expected.
A missing system.ntp.enabled UCI node is required for the bug to show
up. The reasons for why it would be missing in the first place were not
investigated.
Signed-off-by: Henrique de Moraes Holschuh <henrique@nic.br>
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com> [PKG_RELEASE increase]
(cherry picked from commit 556b8581a1)
ab7a39a umdns: fix unused error
45c4953 dns: explicitly endian-convert all fields in header and question
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 22ae8bd50e)
(cherry picked from commit 17c4593e63f5847868f2c38185275199d37d379a)
gcc 8 & 9 appear to be more picky with regards access alignment to
packed structures, leading to this warning in dns.c:
dns.c:261:2: error: converting a packed ‘struct dns_question’ pointer
(alignment 1) to a ‘uint16_t’ {aka ‘short unsigned int’} pointer
(alignment 2) may result in an unaligned pointer value
[-Werror=address-of-packed-member]
261 | uint16_t *swap = (uint16_t *) q;
Work around what I think is a false positive by turning the warning off.
Not ideal, but not quite as not ideal as build failure.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
(cherry picked from commit 02640f0147)
(cherry picked from commit a10b6ec1c8cd6d14a3b76a2ec3d81442b85f7321)
Add option 'scriptarp' to uci dnsmasq config to enable --script-arp functions.
The default setting is false, meaning any scripts in `/etc/hotplug.d/neigh` intended
to be triggered by `/usr/lib/dnsmasq/dhcp-script.sh` will fail to execute.
Also enable --script-arp if has_handlers returns true.
Signed-off-by: Jordan Sokolic <oofnik@gmail.com>
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com> [PKG_RELEASE increase]
This reverts commit 6b7eeb74db since it
didn't contain a reference to the CVE it addresses. The next commit
will re-add the commit including a CVE reference in its commit message.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
8d45443bb5c9 pppd: Ignore received EAP messages when not doing EAP
8d7970b8f3db pppd: Fix bounds check in EAP code
858976b1fc31 radius: Prevent buffer overflow in rc_mksid()
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 215598fd03)
The $(space) definition in the hostapd Makefile ceased to work with
GNU Make 4.3 and later, leading to syntax errors in the generated
Kconfig files.
Drop the superfluous redefinition and reuse the working $(space)
declaration from rules.mk to fix this issue.
Fixes: GH#2713
Ref: https://github.com/openwrt/openwrt/pull/2713#issuecomment-583722469
Reported-by: Karel Kočí <cynerd@email.cz>
Suggested-by: Jonas Gorski <jonas.gorski@gmail.com>
Tested-by: Shaleen Jain <shaleen@jain.sh>
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 766e778226)
hostapd sets minimum values for CWmin/CWmax/AIFS and maximum for TXOP.
The code for applying those values had a few bugs leading to bogus values,
which caused significant latency and packet loss.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
set noscan also for IBSS and remove redundant/obsolete variable.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
(cherry picked from commit 702c70264b)
This is a backport from the dnsmasq master which should fix a bug which
could cause a crash in dnsmasq.
I saw the following crashes in my log:
[522413.117215] do_page_fault(): sending SIGSEGV to dnsmasq for invalid read access from 2a001450
[522413.124464] epc = 004197f1 in dnsmasq[400000+23000]
[522413.129459] ra = 004197ef in dnsmasq[400000+23000]
This is happening in blockdata_write() when block->next is
dereferenced, but I am not sure if this is related to this problem or if
this is a different problem. I am unable to reproduce this problem.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 414d054138)
Enables radio resource management to be reported by hostapd to clients.
Ref: https://github.com/lede-project/source/pull/1430
Co-developed-by: Lorenzo Santina <lorenzo.santina@edu.unito.it>
Signed-off-by: Lorenzo Santina <lorenzo.santina@edu.unito.it>
Signed-off-by: Kyle Copperfield <kmcopper@danwin1210.me>
[removed the DMARC crap]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit 87f9292300)
hostapd will not use the getrandom() syscall and as a fallback use
/dev/random, the syscall is supported since Linux 3.17 and in the musl,
glibc and uclibc version used by OpenWrt.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 998686364d)
All the content of this function is proceeded by IEEE8021X_EAPOL no code
accesses the ssid variable outside of this ifdef.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 0d86bf518a)
Instead of patching the workaround away, just use the config option.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 9b4a27455c)
The size of the ipkgs increase a bit (between 0.7% and 1.1%):
old 2019-04-21 (2.8):
288264 wpad-basic_2019-04-21-63962824-1_mipsel_24kc.ipk
256188 wpad-mini_2019-04-21-63962824-1_mipsel_24kc.ipk
427475 wpad-openssl_2019-04-21-63962824-1_mipsel_24kc.ipk
423071 wpad-wolfssl_2019-04-21-63962824-1_mipsel_24kc.ipk
new 2019-08-08 (2.9):
290217 wpad-basic_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk
258745 wpad-mini_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk
431732 wpad-openssl_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk
427641 wpad-wolfssl_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 167028b750)
This also syncs the configuration files with the default configuration
files, but no extra options are activated or deactivated.
The mesh patches were partially merged into hostapd 2.8, the remaining
patches were extracted from patchwork and are now applied by OpenWrt.
The patches still have open questions which are not fixed by the author.
They were taken from this page:
https://patchwork.ozlabs.org/project/hostap/list/?series=62725&state=*
The changes in 007-mesh-apply-channel-attributes-before-running-Mesh.patch
where first applied to hostapd, but later reverted in hostapd commit
3e949655ccc5 because they caused memory leaks.
The size of the ipkgs increase a bit (between 1.3% and 2.3%):
old 2018-12-02 (2.7):
283337 wpad-basic_2018-12-02-c2c6c01b-11_mipsel_24kc.ipk
252857 wpad-mini_2018-12-02-c2c6c01b-11_mipsel_24kc.ipk
417473 wpad-openssl_2018-12-02-c2c6c01b-11_mipsel_24kc.ipk
415105 wpad-wolfssl_2018-12-02-c2c6c01b-11_mipsel_24kc.ipk
new 2019-04-21 (2.8):
288264 wpad-basic_2019-04-21-63962824-1_mipsel_24kc.ipk
256188 wpad-mini_2019-04-21-63962824-1_mipsel_24kc.ipk
427475 wpad-openssl_2019-04-21-63962824-1_mipsel_24kc.ipk
423071 wpad-wolfssl_2019-04-21-63962824-1_mipsel_24kc.ipk
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Tested-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
(cherry picked from commit 8af79550e6)
For AP mode, OpenWrt automatically sets ieee80211w to either 1 or 2, depending
on whether the encryption is set to sae-mixed, or sae/owe/eap suite-b.
Mirror the same defaults for client mode connections, in order to allow an
OpenWrt station to associate to an OpenWrt ap with SAE, OWE or Suite-B encryption
without the need to manually specify "option ieee80211w" on the station.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit abb4f4075e)
This changes fixes the generation of the wpa_supplicant client configuration
in WPA3 OWE client mode. Instead of incorrectly emitting key_mgmt=NONE, use
the proper key_mgmt=OWE setting instead.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
(cherry picked from commit 4209b28d23)
wps_supplicant.h assumes that 'struct wpa_bss' is forward declared if
CONFIG_WPS is not defined. With the later inclusion of
600-ubus_support, the issue manifests in warnings like these:
wps_supplicant.h:113:15: warning: 'struct wpa_bss' declared inside parameter list will not be visible outside of this definition or declaration
struct wpa_bss *bss)
^~~~~~~
This patch forward declares 'struct wpa_bss' regardless.
Signed-off-by: Leon M. George <leon@georgemail.eu>
[commit message facelift]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit f974f8213b)
The original wpa_hexdump uses a 'void *' for the payload. With patch
410-limit_debug_messages, the signature changes and compiler warnings
occur at various places. One such warning is:
wpa_debug.h:106:20: note: expected 'const u8 * {aka const unsigned char *}' but argument is of type 'struct wpa_eapol_key *'
Signed-off-by: Leon M. George <leon@georgemail.eu>
[commit message facelift]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit a123df2758)
This edjusts the selection of recently removed wolfssl options which
have always been built into the library even in their abscence.
Also remove the selection of libwolfssl itself, allowing the library to
be built as a module.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
(cherry picked from commit 94d131332b)
This enables PMKSA and opportunistic key caching by default for
WPA2/WPA3-Personal, WPA3-Personal and OWE auth types.
Otherwise, Apple devices won't connect to the WPA3 network.
This should not degrade security, as there's no external authentication
provider.
Tested with OCEDO Koala and iPhone 7 (iOS 13.1).
Signed-off-by: David Bauer <mail@david-bauer.net>
(cherry picked from commit 3034f8c3b8)
Signed-off-by: David Bauer <mail@david-bauer.net>
This adds the key_type and ec_curve options to enable the generation of
EC keys during initialization, using openssl or the new options added to
px5g.
Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
(cherry picked from commit 7f2b230b3b)
This commit will activate CONFIG_IEEE80211W for all, but the mini
variant when at least one driver supports it. This will add ieee80211w
support for the mesh variant for example.
Fixes: FS#2397
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
(cherry picked from commit 1d4df52c21)
Daniel Golle
Hauke Mehrtens
Jason A. Donenfeld
Hans Dedecker
Jo-Philipp Wich
Antonio Quartulli
Felix Fietkau
Kevin Darbyshire-Bryant
Henrique de Moraes Holschuh
Jordan Sokolic
Petr Štetiar
Kyle Copperfield
Leon M. George
Eneas U de Queiroz
David Bauer