Mathias Kresin [Sat, 28 Jan 2017 15:19:35 +0000 (16:19 +0100)]
base-files: don't overwrite model name set by target
The condition is always true due to the literal string followed the
-n test parameter. A model name set by target scripts always gets
overwritten this way.
Change the condition to check for an already existing destination file
as it was before 5e85ae9 ("base-files: fix error message during boot").
Sergey Sergeev [Sun, 29 Jan 2017 16:01:40 +0000 (19:01 +0300)]
ledtrig-netdev: don't cancel work on events for different interfaces
This fixes logic bug(in function netdev_trig_notify) introduced in 0b2991a8ed32b5d9ba0a7c3441a72e1c1a1c4fde commit.
Events triggered by different interfaces were stopping work queue so it
wasn't working for tx/rx mode.
HTB and TBF are the basic traffic shapers used by sqm-scripts. Moving
these into kmod-sched-core enables sqm-scripts to downgrade its
dependency from kmod-sched to kmod-sched-core, potentially making it
useful on devices with smaller flash sizes.
This adds around 30k to the size of kmod-sched-core (20k for sch_htb.ko
and 10k for sch_tbf.ko).
Hauke Mehrtens [Fri, 27 Jan 2017 22:36:45 +0000 (23:36 +0100)]
openssl: update to version 1.0.2k
This fixes the following security problems:
CVE-2017-3731: Truncated packet could crash via OOB read
CVE-2017-3732: BN_mod_exp may produce incorrect results on x86_64
CVE-2016-7055: Montgomery multiplication may produce incorrect results
Mathias Kresin [Sat, 28 Jan 2017 11:36:15 +0000 (12:36 +0100)]
lantiq: fix brnImage signature for the VGV7510KW22BRN images
The VGV7510KW22BRN and VGV7519BRN do not have the same brnImage
signature. It was accidentally changed with ba42c1d ("lantiq: un-macro
the image building code").
base-files: fix user creation on sysupgrade with few opkg control files
If only a single opkg control file exists (which can happen with
CONFIG_CLEAN_IPKG), grep would not print the file name by default. Instead
of forcing it using -H, we just switch to -l (print only file names) and
get rid of the cut.
Add -s to suppress an error message when no control files exist.
include/rootfs.mk: keep Require-User lines with CONFIG_CLEAN_IPKG
Require-User is handled by /etc/uci-defaults/13_fix_group_user on first
boot, so we need to keep these when removing all opkg data with
CONFIG_CLEAN_IPKG.
Felix Fietkau [Wed, 25 Jan 2017 14:32:18 +0000 (15:32 +0100)]
ath9k: add stability fixes for long standing hang issues (FS#13, #34, #373, #383)
The radio would stop communicating completely. This issue was easiest to
trigger on AR913x devices, e.g. the TP-Link TL-WR1043ND, but other
hardware was occasionally affected as well.
The most critical issue was a race condition in disabling/enabling IRQs
between the IRQ handler and the IRQ processing tasklet
Felix Fietkau [Tue, 24 Jan 2017 14:23:30 +0000 (15:23 +0100)]
Add back the commit "ath9k: Add airtime fairness scheduler"
This reverts commit c296ba834db4ce8c71e0ad7030aab188fe60b27b.
According to several reports, the issues with the airtime fairness
changes are gone in current versions.
It's time to re-apply the patch now.
Felix Fietkau [Tue, 24 Jan 2017 11:25:05 +0000 (12:25 +0100)]
ramips/rt288x: switch CPU_TYPE to 24kc
While rt288x only has a MIPS 4KEc processor, it implements the MIPS32r2
architecture just like the 24Kc, so the instruction set should be 100%
compatible.
Switching it to 24kc allows it to share the package architecture with a
lot of other targets instead of creating a special case, saving
buildbot resources.
Jo-Philipp Wich [Mon, 23 Jan 2017 13:52:27 +0000 (14:52 +0100)]
hostapd: fix stray "out of range" shell errors in hostapd.sh
The hostapd_append_wpa_key_mgmt() procedure uses the possibly uninitialized
$ieee80211r and $ieee80211w variables in a numerical comparisation, leading
to stray "netifd: radio0 (0000): sh: out of range" errors in logread when
WPA-PSK security is enabled.
Ensure that those variables are substituted with a default value in order to
avoid emitting this (harmless) shell error.
kmod-sched-cake: fix parameter passing kernel/user space
The last two parameters passed between user space tc and kernel space
sched-cake were transposed due to a merge mistake in a parameter header
file.
As such, using a packet overhead figure was likely to set cake to wash
packet DSCP values. Similarly, the DSCP wash flag was used as an offset
to the displayed packet overhead value.
Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
Felix Fietkau [Thu, 19 Jan 2017 13:25:09 +0000 (14:25 +0100)]
lantiq: fix an ethernet stability issue triggered by receving packets during boot
Disabling ethernet during reboot (only to enable it again when the
ethernet driver attaches) can put the chip into a faulty state where it
corrupts the header of all incoming packets.
This happens if packets arrive during the time window where the core is
disabled, and it can be easily reproduced by rebooting while sending a
flood ping to the broadcast address.
Mirko Vogt [Wed, 18 Jan 2017 13:36:45 +0000 (14:36 +0100)]
git-kernel: $(SUBDIR) should always be $(LINUX_VERSION)
Before SUBDIR was set to $(PATCHVER) which may
or may not include the minor version number of
the linux kernel version. Usually it doesn't.
So the git-clone'd linux kernel was packed without
the minor version number taken into account, which
broke further processing, as it expected the
extracted dir being named linux-$(LINUX_VERSION)
(=with minor version) rather than linux-$(PATCHVER)
(=without minor version).
Changing SUBDIR to $(LINUX_VERSION) creates
consistent behaviour here.
Jo-Philipp Wich [Wed, 18 Jan 2017 02:15:31 +0000 (03:15 +0100)]
sdk: do not strip static libraries
Do not strip static libraries shipped with the SDK in order to preserve the
archive index. If we strip the index of the shipped libraries, host programs
will fail to link these libraries with errors like:
libssl.a: error adding symbols: Archive has no index; run ranlib to add one
The error was found while investigating a Python host build failure within
the SDK environment.
base-files: add ARCH_PACKAGES to openwrt_release and os-release
Knowing the package architecture at runtime can be useful, e.g. to
configure opkg repository URLs. The value of ARCH_PACKAGES ("%A" in
VERSION_SED) as added to openwrt_release (as DISTRIB_ARCH) and os-release
(as LEDE_ARCH).
Felix Fietkau [Sat, 14 Jan 2017 21:29:05 +0000 (22:29 +0100)]
kernel: make ledtrig-netdev use a work queue for updates
This fixes hangs in igb that happen if the update call interrupts an
already existing dev_get_stats call. In that case the calling CPU
deadlocks because it's trying to acquire the same spinlock recursively.
Rafał Miłecki [Sun, 15 Jan 2017 22:04:18 +0000 (23:04 +0100)]
mac80211: use wiphy_read_of_freq_limits in brcmfmac
This makes use of cfg80211 feature backported & described in 188626f17c09 ("mac80211: backport cfg80211 support for
ieee80211-freq-limit DT property").
This specifies wireless limitations for Netgear R8000 making sure people
won't use channels incorrectly (limited performance). There are also 2
new DTS files not used by us yet.
Pavel Kubelun [Sat, 14 Jan 2017 14:01:09 +0000 (17:01 +0300)]
ipq806x: refactor ipq8065 device tree
At the moment we have 2 seperate device tree sources: 1 for ipq8064 soc located
in kernel source and 1 custom in LEDE for ipq8065.
ipq8064 and ipq8065 SoCs are completely identical except ipq8065 has higher cpu
frequencies and adjusted corresponding power supply.
This commit makes ipq8065 DTS to contain only specific for ipq8065 DT entries while
pulling all the basic SoC stuff from ipq8064 DTS.
It makes easier to manage ipq806x device trees and instead of committing changes into
2 seperate equal DTS we are ending up with only 1.
It also enables ipq8065 devices to automatically receive changes made in upstream kernel.
Signed-off-by: Pavel Kubelun <be.dissent@gmail.com>
Pavel Kubelun [Fri, 23 Dec 2016 22:51:07 +0000 (01:51 +0300)]
ipq806x: disable usb3 phy suspend and add usb tcsr control
According to Qualcomm SDK usb3 phy suspend should be disabled for
ipq806x.
This may solve issue on some ipq806x devices that breaks usb3
storage peripherals during system boot.
Signed-off-by: Pavel Kubelun <be.dissent@gmail.com>
Lazar Demin [Mon, 9 Jan 2017 19:59:56 +0000 (19:59 +0000)]
ramips: Added Onion Omega2 and Omega2+
This patch adds support for the Onion Omega2 and Omega2+ (https://onion.io)
Specifications:
- SoC: MediaTek MT7688AN (580MHz, ramips)
- Omega2
- RAM: 64MB DDR
- Storage: 16MB NOR SPI flash onboard
- Omega2+
- RAM: 128MB DDR
- Storage: 32MB NOR SPI flash onboard + microSD slot
- Wireless: Built into MT7688AN (mt76) with onboard 1x chip antenna and u.FL connecter
- Ethernet: 1x100M pins on Omega2 & Omega2+, can use Ethernet Expansion and an Omega Dock to get a physical Ethernet port
- Strongly recommend using the Omega2 & Omega2+ with a Dock (Expansion Dock, Power Dock, Arduino Dock 2, Mini Dock)
- All Docks Provide:
- Micro-USB port to provide power to the Omega
- On the Expansion and Mini Docks, can also access the terminal (UART0) via serial
- USB 2.0 socket connected to Omega
- Just the Expansion Dock, Power Dock, and Arduino Dock 2 provide:
- Omega GPIO breakout
- Allows for connection of Omega Expansions:
- Ethernet Expansion
- Relay Expansion
- PWM Expansion
- OLED Expansion
- Ethernet Expansion
- Proto Expansion
- Cellular Expansion
Rosen Penev [Fri, 13 Jan 2017 15:14:38 +0000 (07:14 -0800)]
ramips: Add I2C driver to the default kernel config
I made a commit that added the RTC driver to the kernel config with
the intent that it would fix hctosys. Unfortunately while the RTC
driver is in there, it's connected through I2C, the driver for which
comes in module form and is thus loaded late. After this commit, it
works fine.
Felix Fietkau [Wed, 11 Jan 2017 09:57:49 +0000 (10:57 +0100)]
kernel: backport MIPS changes introducing a separate IRQ stack
Prevents crashes when IRQs arrive when the current kernel stack context
already contains deeply nested function calls, e.g. when stacking lots
of network devices on top of each other
Jo-Philipp Wich [Fri, 13 Jan 2017 17:31:08 +0000 (18:31 +0100)]
firewall: fix forwarding local subnet traffic
Packets which are merely forwarded by the router and which are neither
involved in any DNAT/SNAT nor originate locally, are considered INVALID
from a conntrack point of view, causing them to get dropped in the
zone_*_dest_ACCEPT chains, since those only allow stream with state NEW
or UNTRACKED.
Remove the ctstate restriction on dest accept chains to properly pass-
through unrelated 3rd party traffic.
projects / thirdparty / openwrt.git / log
