The implementation is not efficient on ar8xxx switches. It triggers high
CPU load and degrades device performance.
The high CPU load has been traced down to the ar8xxx_reg_wait() call in
ar8xxx_mib_op(), which has to usleep_range() till the MIB busy flag set
by the request to update the MIB counter is cleared.
This commit removes the get_port_stats() code introduced in 4d8a66d and
leaves a note for future hacker's beware.
Fixes: FS#1004
Signed-off-by: Thibaut VARENE <hacks@slashdirt.org>
Fix multiple syntax errors in shelscripts (of packages only)
These errors were causing many conditions to not working properly
Signed-off-by: Lorenzo Santina <lorenzo.santina@edu.unito.it>
[increase PKG_RELEASE, drop command substitution from directip.sh]
Signed-off-by: Mathias Kresin <dev@kresin.em>
Set sysctl fs.suid_dumpable = 2
This allows suid processes to dump core according to kernel.core_pattern
setting. LEDE typically uses suid to drop root priviledge rather than
gain it but without this setting any suid process would be unable to
produce coredumps (e.g. dnsmasq)
Processes still need to set a non zero core file process limit ('ulimit
-c unlimited' or if procd used 'procd_set_param limits
core="unlimited"') in order to produce a core. This setting removes an
obscure stumbling block along the way.
>From https://www.kernel.org/doc/Documentation/sysctl/fs.txt
suid_dumpable:
This value can be used to query and set the core dump mode for setuid
or otherwise protected/tainted binaries. The modes are
0 - (default) - traditional behaviour. Any process which has changed
privilege levels or is execute only will not be dumped.
1 - (debug) - all processes dump core when possible. The core dump is
owned by the current user and no security is applied. This is
intended for system debugging situations only. Ptrace is unchecked.
This is insecure as it allows regular users to examine the memory
contents of privileged processes.
2 - (suidsafe) - any binary which normally would not be dumped is dumped
anyway, but only if the "core_pattern" kernel sysctl is set to
either a pipe handler or a fully qualified path. (For more details
on this limitation, see CVE-2006-2451.) This mode is appropriate
when administrators are attempting to debug problems in a normal
environment, and either have a core dump pipe handler that knows
to treat privileged core dumps with care, or specific directory
defined for catching core dumps. If a core dump happens without
a pipe handler or fully qualifid path, a message will be emitted
to syslog warning about the lack of a correct setting.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com> [PKG_RELEASE increase]
The addresses were read from the 'config' partition, which would not always
contain the addresses at the same offsets, depending on the stock firmware
version used before flashing LEDE. Change this to get the addresses from
the 'product-info' partition, which is read-only.
Reported-and-tested-by: Andreas Ziegler <ml@andreas-ziegler.de>
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
The ZBT WE1026-5G
(http://www.zbtlink.com/products/router/WE1026-5G.html) is the follow-up
to the ZBT WE1026 and is based on MT7620. For the previous WE1026, the
ZBT WE826 image could be used. However, as the name implies, the -5G
comes equipped with a 5GHz wifi radio. As the WE826 only has a 2.4GHz
radio, the addition of 5GHz means that a separate image is needed for
the WE1026-5G. I suspect that this image will also work on the previous
WE1026, but I don't have a device to test with.
The WE1026-5G has following specifications:
* CPU: MT7620A
* 1x 10/100Mbps Ethernet.
* 16 MB Flash.
* 64 MB RAM.
* 1x USB 2.0 port.
* 1x mini-PCIe slots.
* 1x SIM slots.
* 1x 2.4Ghz WIFI.
* 1x 5GHz wifi (MT7612)
* 1x button.
* 3x controllable LEDs.
Works:
* Wifi.
* Switch.
* mini-PCIe slot. Only tested with a USB device (a modem).
* SIM slot.
* Sysupgrade.
* Button (reset).
Not working:
* The 5GHz WIFI LED is completely dead. I suspect the issue is the same
as on other devices with Mediatek 5Ghz wifi-cards/chips. The LED is
controlled by the driver, and mt76 (currently) does not support this.
Not tested:
* SD card reader.
Notes:
* The modem (labeled 3G/4G) and power LEDs are controlled by the
hardware.
* There is a 32MB version of this device available, but I do not have
access to it. I have therefor only added support for the 16MB version,
but added all the required infrastructure to make adding support for the
32MB version easy.
Installation:
The router comes pre-installed with OpenWRT, including a variant of
Luci. The initial firmware install can be done through this UI,
following normal procedure. I.e., access the UI and update the firmware
using the sysupgrade-image. Remember to select that you do not want to
keep existing settings.
Recovery:
If you brick the device, the WE1026-5G supports recovery using HTTP. Keep the
reset button pressed for ~5sec when booting to start the web server. Set the
address of the network interface on your machine to 192.168.1.2/24, and
point your browser to 192.168.1.1 to access the recovery UI. From the
recovery UI you can upload a firmware image.
Signed-off-by: Kristian Evensen <kristian.evensen@gmail.com>
The upstream ath10k driver disables the intermediate softqueues for some
devices. This patch reverts that behaviour and always enables the
softqueues (and associated bufferbloat fixes). We have had reports of people
running this with good results:
https://lists.bufferbloat.net/pipermail/make-wifi-fast/2017-September/001497.html
This also refreshes mac80211 patches.
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Currently, if the provided hash is unsupported (length different from 32
or 64 bytes), we happily download the requested file without any kind of
checksum verification.
This is quite dangerous and may provide a false sense of security, because
a single typo in the hash (e.g. one character deleted by mistake) may skip
checksum verification entirely.
Instead, fail immediately if we don't support the provided hash.
In particular, if an external package repository decides to change the
hash algorithm one day, we will now fail loudly instead of skipping
checksum verification without complaints.
Note: if some users of scripts/download.pl knowingly provide an empty hash
because they don't need checksum verification, this change will break
them. This does not seem to be the case currently, but if this feature is
ever needed, an option should be added to download.pl instead of relying
on the hash being empty.
Fixes: eaa4eba10a ("scripts/download.pl: add SHA-256 support")
Signed-off-by: Baptiste Jonglez <git@bitsofnetworks.org>
On a TL-WN710N, this patch increases iperf performance from ~92.5 to ~93.5 mbps. Keep in mind the WN710N is a 100mbps device. I expect greater numbers from gigabit devices.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
The cns3xxx interrupt controller uses a single register and as such
the 'mask' reg/functions must be used as opposed to the 'enable'/'disable'
reg/functions.
This fixes an issue that occurs if more than one GPIO on a specific controller
(there is GPIOA and GPIOB each having 32 GPIO's) uses interrupts. When one
would get enabled all others would be disabled prior to this patch.
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
Acked-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Refresh patches.
Compile-tested on octeon and x86/64.
Runtime-tested on octeon and x86/64.
Fixes CVE-2017-11600.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Fixed an authentication bypass issue in SSL/TLS. When the TLS
authentication mode was set to 'optional',
mbedtls_ssl_get_verify_result() would incorrectly return 0 when the
peer's X.509 certificate chain had more than
MBEDTLS_X509_MAX_INTERMEDIATE_CA intermediates (default: 8), even when
it was not trusted. This could be triggered remotely on both the client
and server side. (Note, with the authentication mode set by
mbedtls_ssl_conf_authmode()to be 'required' (the default), the handshake
was correctly aborted).
Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
Tested-by: Magnus Kroken <mkroken@gmail.com>
This adds metadata to wpj344 and wpj558 images to prevent loading
firmware of wpj344 into wpj558 and vice versa. This until now was
possible and break the units and had to be recovered from the uboot.
Signed-off-by: Enrique Giraldo <enrique.giraldo@galgus.net>
COMFAST CF-E355AC is a ceiling mount AP with PoE support, based on
Qualcomm/Atheros QCA9531 + QCA9882.
Short specification:
- 2x 10/100 Mbps Ethernet, with PoE support
- 64MB of RAM (DDR2)
- 16 MB of FLASH
- 2T2R 2.4 GHz, 802.11b/g/n
- 2T2R 5 GHz, 802.11ac/n/a
- built-in 4x 3 dBi antennas
- output power (max): 500 mW (27 dBm)
- 1x RGB LED, 1x button
- built-in watchdog chipset
Flash instruction:
Original firmware is based on OpenWrt.
Use sysupgrade image directly in vendor GUI.
Signed-off-by: Enrique Giraldo <enrique.giraldo@galgus.net>
[whitespace fixes, ac radio caldata offset fix]
Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
GL.iNet GL-USB150 is an USB dongle WiFi router, based on Atheros AR9331.
Specification:
- 400/400/200 MHz (CPU/DDR/AHB)
- 64 MB of RAM (DDR2)
- 16 MB of FLASH (SPI NOR)
- Realtek RTL8152B USB to Ethernet bridge (connected with AR9331 PHY4)
- 1T1R 2.4 GHz
- 2x LED, 1x button
- UART header on PCB
Flash instruction:
Vendor firmware is based on OpenWrt CC. GUI or sysupgrade can be used
to flash LEDE firmware.
Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
ifname variable were not assigned due to syntax error
causing the hostapd config file to have an empty iapp_interface= option
Signed-off-by: Lorenzo Santina <lorenzo.santina.dev@gmail.com>
For the Archer C50v1, the EU and US versions are differentiated by their
respective HW additional version (0x0 for US, 0x2 for EU).
The stock web interface checks this field before flashing, making it
impossible to flash the current (US) factory image on EU hardware.
However the bootloader does not check this field, making it possible to use
a single sysupgrade image for both hardware.
This patch adds the necessary build bits to generate both EU and US factory
images, and renames the target as "Archer C50v1" since there are as of now
3 different versions of Archer C50 (all with different CPUs).
Signed-off-by: Thibaut VARENE <hacks@slashdirt.org>
The HNET C108
(http://www.szhwtech88.com/Product-product-cid-100-id-4374.html) is a
mifi based on MT7602A, which has the following specifications:
* CPU: MT7620A
* 1x 10/100Mbps Ethernet.
* 16 MB Flash.
* 64 MB RAM.
* 1x USB 2.0 port. Only power is connected, this port is meant for
charging other devices.
* 1x mini-PCIe slots.
* 1x SIM slots.
* 1x 2.4Ghz WIFI.
* 1x button.
* 6000 mAh battery.
* 5x controllable LEDs.
Works:
* Wifi.
* Switch.
* mini-PCIe slot. Only tested with a USB device (a modem).
* SIM slot.
* Sysupgrade.
* Button (reset).
Not working (also applies to the factory firmware):
* Wifi LED. It is always switched on, there is no relation to the
up/down state or activity of the wireless interface.
Not tested:
* SD card reader.
Notes:
* The C108 has no dedicated status LED. I therefore set the LAN LED as
status LED.
Installation:
The router comes pre-installed with OpenWRT, including a variant of
Luci. The initial firmware install can be done through this UI,
following normal procedure. I.e., access the UI and update the firmware
using the sysupgrade-image. Remember to select that you do not want to
keep existing settings.
Recovery:
If you brick the device, the C108 supports recovery using TFTP. Keep the
reset button pressed for ~5sec when booting to trigger TFTP. Set the
address of the network interface on your machine to 10.10.10.3/24, and
rename your image file to Kernal.bin.
Signed-off-by: Kristian Evensen <kristian.evensen@gmail.com>
In case the link changes from down to up, the register is only updated
on read. If the link failed/was down, this bit will be 0 until after
reading this bit again.
Fixes a reported link down by swconfig alebit the link is up (query for
the link again will show the correct link status)
Signed-off-by: Mathias Kresin <dev@kresin.me>
In conditions where none of the switch ports is connected during boot,
the priv->port[i].link != priv->port[i].phydev->link condition is false
since both link values are equal (false). The carrier of the switch
netdev is never set to off and the link state reported by ip is UNKNOWN.
Turn the carrier off if none of the switch ports has a link, regardless
whether something has been changed. Add a check for a carrier to
prevent unnecessary calls to netif_carrier_off() if the carrier is
already off.
Based on a patch send by Martin Schiller.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Don't return arcount=1 if EDNS0 RR won't fit in the packet.
Omitting the EDNS0 RR but setting arcount gives a malformed packet.
Also, don't accept UDP packet size less than 512 in received EDNS0.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Remove LEDE partial fix for CVE-2017-13704.
Backport official fix from upstream.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com> (PKG_RELEASE increase)
Commit 77645ffcd9 ("ramips: add support for the GnuBee Personal Cloud
One") dropped the execution permission from 01_leds with the result
that the file isn't started during first boot and no default LED
configuration is added.
Revert the introduced file permission change.
Fixes: FS#979
Signed-off-by: Kristian Evensen <kristian.evensen@gmail.com>
[cherry picked the fix from a board support patch]
Signed-off-by: Mathias Kresin <dev@kresin.me>
Extendprefix is typically used to extend an IPv6 RA prefix from a mobile
wan link to the LAN; such scenario requires correct RA prefix settings
like the on link flag not being set.
However some mobile manufacter set the RA prefix on link flag which breaks
basic IPv6 routing.
Work around this issue by filtering out the route being equal to the
extended prefix.
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
The WAN port on the Netgear WNDR4300 router has two LEDs,
amber and green. Use the switch LED trigger to behave as the
rest of the LAN HW controlled LEDs
- Green: 1 Gbps
- Amber: 100/10 Mbps
Signed-off-by: Daniel Gonzalez Cabanelas <dgcbueu@gmail.com>
1. Add support to LAN/WAN LEDs attached to ar8327.
2. Fix the problem that LAN/WAN LEDs does not blink in hardware (auto)
mode when connected to 10M/100M ethernet.
Signed-off-by: Kuang Rufan <master@a1983.com.cn>
51733a6 ra: align RA update interval with RFC4861 (FS#964)
Add ra_holdoff config option which allows to configure the RA minimum
update interval which is by default 3 seconds as stated in RFC4861.
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
Refresh patches.
Compile-tested on ramips/mt7621 and x86/64.
Runtime-tested on ramips/mt7621 and x86/64.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
b1bc8d5 kmodloader: log error message in case of out of memory
f346111 kmodloader: lift restriction on module alias info
f1ef2c3 kmodloader: fix possible segfaults
9cb63df kmodloader: fix endianess check
2cff779 kmodloader: Check module endian before loading
d54f38a kmodloader/get_module_info: initialized aliases to make it more clean
a0b6fef kmodloader: insmod: fix a memoryleak in error case
278c4c4 kmodloader/get_module_name: null-terminate the string
16f7e16 syslog: remove unnecessary sizeof struct between messages
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
This generic structure defines tx_bytes and rx_bytes as unsigned long (u32),
while several devices would typically report unsigned long long (u64).
The code can work as is, but there's a chance that with a sufficiently fast
interface the overflow might happen too fast to be correctly noticed by the
consumers of this data.
This patch makes both field unsigned long long and updates the only known
consumer of this data: swconfig_leds.c
Signed-off-by: Thibaut VARENE <hacks@slashdirt.org>
This patch provides a generic switch_dev_ops 'get_port_stats()' callback by
taping into the relevant port MIB counters.
This callback is used by swconfig_leds led trigger to blink LEDs with port
network traffic.
Signed-off-by: Thibaut VARENE <hacks@slashdirt.org>
This patch provides a generic switch_dev_ops 'get_port_stats()' callback by
taping into the relevant port MIB counters.
This callback is used by swconfig_leds led trigger to blink LEDs with port
network traffic.
Signed-off-by: Thibaut VARENE <hacks@slashdirt.org>
