openwrt/openwrt.git
6 years agofirewall: bump to git HEAD
Stijn Tintel [Fri, 2 Mar 2018 10:14:11 +0000 (11:14 +0100)]
firewall: bump to git HEAD

392811a ubus: let fw3_ubus_address() return the number of resolved addresses
359adcf options: emit an empty address item when resolving networks fails
503db4a zones: disable masq when resolving of all masq_src or masq_dest items failed
f50a524 helpers: implement explicit CT helper assignment support
a3ef503 zones: allow per-table log control
8ef12cb iptables: fix possible NULL pointer access on constructing rule masks

Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
6 years agobase-files: tune fragment queue thresholds for available system memory
Matthias Schiffer [Mon, 5 Mar 2018 08:13:53 +0000 (09:13 +0100)]
base-files: tune fragment queue thresholds for available system memory

The default fragment low/high thresholds are 3 and 4 MB. On devices with
only 32MB RAM, these settings may lead to OOM when many fragments that
cannot be reassembled are received. Decrease fragment low/high thresholds
to 384 and 512 kB on devices with less than 64 MB RAM.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agobcm53xx: include patch adding WZR-900DHP LEDs support
Rafał Miłecki [Wed, 7 Mar 2018 16:00:30 +0000 (17:00 +0100)]
bcm53xx: include patch adding WZR-900DHP LEDs support

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agoinclude/package-defaults.mk: fix default Build/Prepare with empty ./src
Matthias Schiffer [Tue, 6 Mar 2018 20:55:01 +0000 (21:55 +0100)]
include/package-defaults.mk: fix default Build/Prepare with empty ./src

Copying ./src/* would fail when src exists, but is empty or only contains
hidden files.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agoinclude/image.mk: base package-list manifest on unprepared rootfs
Matthias Schiffer [Tue, 6 Mar 2018 13:42:37 +0000 (14:42 +0100)]
include/image.mk: base package-list manifest on unprepared rootfs

With CONFIG_CLEAN_IPKG, the prepared rootfs does not have a full opkg
status file.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agoimagebuilder: remove split patch dirs from imagebuilder archive
Matthias Schiffer [Tue, 6 Mar 2018 20:50:05 +0000 (21:50 +0100)]
imagebuilder: remove split patch dirs from imagebuilder archive

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agoimagebuilder: reuse rootfs preparation from rootfs.mk
Matthias Schiffer [Tue, 6 Mar 2018 08:38:29 +0000 (09:38 +0100)]
imagebuilder: reuse rootfs preparation from rootfs.mk

In addition to removing redundant code, this fixes various issues in
IB-generated images that have been fixed in prepare_rootfs before,
including better handling of CONFIG_CLEAN_IPKG and enabling of initscripts
from FILES.

We also reuse the opkg macro and remove --force-... flags that have been
removed from rootfs.mk as well.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agoinclude/rootfs.mk: pass additional files dir to prepare_rootfs as an argument
Matthias Schiffer [Tue, 6 Mar 2018 20:06:30 +0000 (21:06 +0100)]
include/rootfs.mk: pass additional files dir to prepare_rootfs as an argument

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agoinclude/rootfs.mk: retain list of conffiles with CONFIG_CLEAN_IPKG
Matthias Schiffer [Tue, 6 Mar 2018 13:29:46 +0000 (14:29 +0100)]
include/rootfs.mk: retain list of conffiles with CONFIG_CLEAN_IPKG

/usr/lib/opkg/status must not be removed completely, otherwise the
packages' conffile lists will be missing. Replace it with a reduced version
only containing the conffile entries.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agoinclude/rootfs.mk: do not remove opkg prerm scripts during rootfs preparation
Matthias Schiffer [Tue, 6 Mar 2018 07:49:14 +0000 (08:49 +0100)]
include/rootfs.mk: do not remove opkg prerm scripts during rootfs preparation

When a user removes a preinstalled opkg package, the package's prerm script
(and in particular our default_prerm) should run.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agobase-files: sysupgrade: do not rely on opkg to list changed conffiles
Matthias Schiffer [Tue, 6 Mar 2018 07:25:32 +0000 (08:25 +0100)]
base-files: sysupgrade: do not rely on opkg to list changed conffiles

Many packages use the opkg conffiles field to list configuration files that
are to be retained on upgrades. Make this work on systems without opkg.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agolantiq/xrx200: add usif support to dts files
Martin Schiller [Fri, 23 Feb 2018 13:15:49 +0000 (14:15 +0100)]
lantiq/xrx200: add usif support to dts files

The Lantiq XRX200 aka VR9 doesn't have an asc0. Instead,
there is an USIF module which can either be an UART or a
SPI Controller.

Signed-off-by: Martin Schiller <ms@dev.tdt.de>
6 years agolantiq: ltq-adsl: deactivate ASLR support
Mathias Kresin [Tue, 6 Mar 2018 18:43:40 +0000 (19:43 +0100)]
lantiq: ltq-adsl: deactivate ASLR support

The package still leaks some user space linker options into the kernel
space. This breaks the build when ASLR is activated, deactivate it for
now.

Signed-off-by: Mathias Kresin <dev@kresin.me>
6 years agoramips: mt7620n: enable port 4 as EPHY by default
Giuseppe Lippolis [Sat, 24 Feb 2018 08:45:12 +0000 (09:45 +0100)]
ramips: mt7620n: enable port 4 as EPHY by default

According to the datasheet the mt7620n have a fixed switch configuration
with 5 ephy (10/100) port. No RGMII configuration is possible.

Drop the mdio node as well. Without RGMII, the mdio node doesn't make any
sense

Signed-off-by: Giuseppe Lippolis <giu.lippolis@gmail.com>
[drop mdio node, enable port4 by default]
Signed-off-by: Mathias Kresin <dev@kresin.me>
6 years agoramips: mt7620: power up ephy port 4
Pawel Dembicki [Mon, 19 Feb 2018 21:15:56 +0000 (22:15 +0100)]
ramips: mt7620: power up ephy port 4

In some boards port 4 of mt7620 switch is powered down in bootloader
configuration as lan/wan package leak mitigation.

Reset ephy port 4 to default settings to power up the port.

Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
[clarify issue in commit message]
Signed-off-by: Mathias Kresin <dev@kresin.me>
6 years agowireguard: bump to 20180304
Jason A. Donenfeld [Mon, 5 Mar 2018 10:31:53 +0000 (11:31 +0100)]
wireguard: bump to 20180304

7c0d711 version: bump snapshot
b6a5cc0 contrib: add extract-handshakes kprobe example
37dc953 wg-quick: if resolvconf/run/iface exists, use it
1f9be19 wg-quick: if resolvconf/interface-order exists, use it
4d2d395 noise: align static_identity keys
14395d2 compat: use correct -include path
38c6d8f noise: fix function prototype
302d0c0 global: in gnu code, use un-underscored asm
ff4e06b messages: MESSAGE_TOTAL is unused
ea81962 crypto: read only after init
e35f409 Kconfig: require DST_CACHE explicitly
9d5baf7 Revert "contrib: keygen-html: rewrite in pure javascript"
6e09a46 contrib: keygen-html: rewrite in pure javascript
e0af0f4 compat: workaround netlink refcount bug
ec65415 contrib: embedded-wg-library: add key generation functions
06099b8 allowedips: fix comment style
ce04251 contrib: embedded-wg-library: add ability to add and del interfaces
7403191 queueing: skb_reset: mark as xnet

Changes:

* queueing: skb_reset: mark as xnet

This allows cgroups to classify packets.

* contrib: embedded-wg-library: add ability to add and del interfaces
* contrib: embedded-wg-library: add key generation functions

The embeddable library gains a few extra tricks, for people implementing
plugins for various network managers.

* crypto: read only after init
* allowedips: fix comment style
* messages: MESSAGE_TOTAL is unused
* global: in gnu code, use un-underscored asm
* noise: fix function prototype

Small cleanups.

* compat: workaround netlink refcount bug

An upstream refcounting bug meant that in certain situations it became
impossible to unload the module. So, we work around it in the compat code. The
problem has been fixed in 4.16.

* contrib: keygen-html: rewrite in pure javascript
* Revert "contrib: keygen-html: rewrite in pure javascript"

We nearly moved away from emscripten'ing the fiat32 code, but the resultant
floating point javascript was just too terrifying.

* Kconfig: require DST_CACHE explicitly

Required for certain frankenkernels.

* compat: use correct -include path

Fixes certain out-of-tree build systems.

* noise: align static_identity keys

Gives us better alignment of private keys.

* wg-quick: if resolvconf/interface-order exists, use it
* wg-quick: if resolvconf/run/iface exists, use it

Better compatibility with Debian's resolvconf.

* contrib: add extract-handshakes kprobe example

Small utility for extracting ephemeral key data from the kernel's memory.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com> (git log --oneline description)
6 years agonghttp2: bump to 1.31.0
Hans Dedecker [Mon, 5 Mar 2018 09:23:09 +0000 (10:23 +0100)]
nghttp2: bump to 1.31.0

6e744662 Update bash_completion
478eac09 Update manual pages
88e2029e Bump up version number to 1.31.0, LT revision to 30:0:16
45d76cf5 nghttpx: Close listening socket on graceful shutdown
54573f28 Merge pull request #1137 from nghttp2/session-set-user-data
17793e99 Add nghttp2_session_set_user_data() public API function
5eac3c90 Update manual pages
e70195ae nghttpx: Update doc
fe51e7fa Merge pull request #1130 from nghttp2/avoid-inet_pton-macro
eb951c2c src: Define nghttp2_inet_pton wrapper to avoid inet_pton macro
39f0ce7c Merge pull request #1126 from nghttp2/nghttpx-expired-client-cert
65157811 Merge pull request #1123 from nghttp2/mruby-client-cert-not-before-after
e8af7afc nghttpx: Add an option to accept expired client certificate
38abfd18 nghttpx: Add mruby tls_client_not_before, and tls_client_not_after
ff3edc09 nghttpx: Fix potential memory leak
0bb15406 Bump up version number to 1.31.0-DEV

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agomt76: update to the latest version
Felix Fietkau [Mon, 26 Feb 2018 21:15:24 +0000 (22:15 +0100)]
mt76: update to the latest version

cda627f mt76: add missing VHT maximum A-MPDU length capability
ffa2069 mt76: fix potential sleep in atomic context

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agokernel: remove an old kernel compatibility line from module packaging
Felix Fietkau [Thu, 22 Feb 2018 11:16:19 +0000 (12:16 +0100)]
kernel: remove an old kernel compatibility line from module packaging

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agoar71xx: fix mii bus irq assignment
Felix Fietkau [Sat, 3 Mar 2018 11:23:51 +0000 (12:23 +0100)]
ar71xx: fix mii bus irq assignment

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agoar71xx: remove pdata->fifo_cfg*
Felix Fietkau [Fri, 2 Mar 2018 20:30:10 +0000 (21:30 +0100)]
ar71xx: remove pdata->fifo_cfg*

The values are the same for all chipsets that use it

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agomac80211: avoid sampling rates that are too slow, improves throughput
Felix Fietkau [Sat, 3 Mar 2018 17:51:49 +0000 (18:51 +0100)]
mac80211: avoid sampling rates that are too slow, improves throughput

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agokernel: bump 4.4 to 4.4.119
Stijn Segers [Fri, 2 Mar 2018 07:31:03 +0000 (08:31 +0100)]
kernel: bump 4.4 to 4.4.119

This bumps the 4.4. kernel in master to 4.4.119.
Includes more Meltdown & Spectre mitigation.

* Refresh patches.
* Refresh x86/config for RETPOLINE.
* Deleted 8049-PCI-layerscape-Add-fsl-ls2085a-pcie-compatible-ID.patch (accepted upstream)
* Deleted 8050-PCI-layerscape-Fix-MSG-TLP-drop-setting.patch (accepted upstream)
* Deleted 650-pppoe_header_pad.patch (does not apply anymore (code was replaced)).

Bumps from 4.4.112 to 4.4.115 were handled by Kevin Darbyshire-Bryant.

Compile-tested on: ar71xx & oxnas.

Signed-off-by: Stijn Segers <foss@volatilesystems.org>
Tested-by: Rosen Penev <rosenp@gmail.com>
6 years agokernel: bump 4.9 to 4.9.85
Magnus Kroken [Sat, 3 Mar 2018 00:12:07 +0000 (01:12 +0100)]
kernel: bump 4.9 to 4.9.85

Signed-off-by: Magnus Kroken <mkroken@gmail.com>
6 years agolantiq: Deactivate ASLR support for some applications
Hauke Mehrtens [Sat, 3 Mar 2018 00:21:08 +0000 (01:21 +0100)]
lantiq: Deactivate ASLR support for some applications

The lantiq components still leak some user space linker options into the
kernel space. This breaks with build when ASLR is activated, deactivate
it for now on these packages.

Fixes: FS#1391
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agolantiq: ltq-vmmc: Do not leak user space CFLAGS into kernel space
Hauke Mehrtens [Fri, 2 Mar 2018 23:52:16 +0000 (00:52 +0100)]
lantiq: ltq-vmmc: Do not leak user space CFLAGS into kernel space

Do not leak the user space CFLAGS into the kernel space any more, this
allows us to activate the MIPS16 build.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agolantiq: ltq-tapi: Do not leak user space CFLAGS into kernel space
Hauke Mehrtens [Fri, 2 Mar 2018 23:51:40 +0000 (00:51 +0100)]
lantiq: ltq-tapi: Do not leak user space CFLAGS into kernel space

Do not leak the user space CFLAGS into the kernel space any more, this
allows us to activate the MIPS16 build.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agolantiq: ltq-adsl: Do not leak user space CFLAGS into kernel space
Hauke Mehrtens [Sat, 3 Mar 2018 00:38:50 +0000 (01:38 +0100)]
lantiq: ltq-adsl: Do not leak user space CFLAGS into kernel space

Do not leak the user space CFLAGS into the kernel space any more, this
allows us to activate the MIPS16 build.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agolantiq: ltq-vdsl: Do not leak user space CFLAGS into kernel space
Hauke Mehrtens [Fri, 2 Mar 2018 23:25:06 +0000 (00:25 +0100)]
lantiq: ltq-vdsl: Do not leak user space CFLAGS into kernel space

Do not leak the user space CFLAGS into the kernel space any more, this
allows us to activate the MIPS16 build.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agolantiq: ltq-vdsl-mei: Do not leak user space CFLAGS into kernel space
Hauke Mehrtens [Fri, 2 Mar 2018 23:23:55 +0000 (00:23 +0100)]
lantiq: ltq-vdsl-mei: Do not leak user space CFLAGS into kernel space

Do not leak the user space CFLAGS into the kernel space any more, this
allows us to activate the MIPS16 build.

This decreases the size of the ipk file from 87589 bytes to 81267 bytes.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agolantiq: ltq-ifxos: Do not force O3 build any more
Hauke Mehrtens [Fri, 2 Mar 2018 23:22:49 +0000 (00:22 +0100)]
lantiq: ltq-ifxos: Do not force O3 build any more

Do not force to build with O3 optimization any more, but take what was
selected by the OpenWrt build system.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agolantiq: ltq-ifxos: activate build with mips16
Hauke Mehrtens [Fri, 2 Mar 2018 23:20:43 +0000 (00:20 +0100)]
lantiq: ltq-ifxos: activate build with mips16

The build process does not leak the user space cflags into the kernel
build process any more, this allows to activate MIPS16 builds.
This was fixed with some update of ifxos.

This decreases size of the  libifxos.a and the ltq-vdsl-app
old:
78320 libifxos.a
44383 ltq-vdsl-app_4.17.18.6-2_mips_24kc.ipk

new:
66852 libifxos.a
43506 ltq-vdsl-app_4.17.18.6-2_mips_24kc.ipk

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agokernel: generic: Add kernel configuration options
Hauke Mehrtens [Fri, 2 Mar 2018 21:32:13 +0000 (22:32 +0100)]
kernel: generic: Add kernel configuration options

These options are needed for the mvebu arm64 target.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agokernel: bump 4.14 to 4.14.23
Stijn Segers [Wed, 28 Feb 2018 21:53:37 +0000 (22:53 +0100)]
kernel: bump 4.14 to 4.14.23

This patch bumps the 4.14 kernel to .23.
- Refreshed patches.
- Deleted bcm53xx/patches-4.14/089-PCI-iproc-Fix-NULL-pointer-dereference-for-BCMA.patch. Has been accepted upstream.
- Deleted generic/pending-4.14/821-usb-Remove-annoying-warning-about-bogus-URB.patch. The upstream URB code was changed,
  the patch no longer applies. I discussed this with the patch author and removed it for now, we'll see how it goes.

Compile-tested on: ramips/mt7621, x86/64
Run-tested on: ramips/mt7621, x86/64

Signed-off-by: Stijn Segers <foss@volatilesystems.org>
6 years agomvebu: base-files: moving on from net-tools
Tomasz Maciej Nowak [Wed, 24 Jan 2018 15:41:32 +0000 (16:41 +0100)]
mvebu: base-files: moving on from net-tools

Remove dependency on net-tools, replacing calls to ifconfig with iproute
equivalent, in functions changing interface MAC addresses.

Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
6 years agomvebu: rename clearfog sysupgrade script and functions
Tomasz Maciej Nowak [Wed, 24 Jan 2018 15:39:48 +0000 (16:39 +0100)]
mvebu: rename clearfog sysupgrade script and functions

Cosmetic name change of ClearFog sysupgrade script and its functions to
more generic sdcard. This way it won't be confusing for other future
device additions.

Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
6 years agomvebu: remove fs declaration in mount commands
Tomasz Maciej Nowak [Wed, 24 Jan 2018 23:04:56 +0000 (00:04 +0100)]
mvebu: remove fs declaration in mount commands

Allows to have other file system for boot partition without breaking
sysupgrade.

Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
6 years agomvebu: unify boot.scr creation
Tomasz Maciej Nowak [Wed, 14 Feb 2018 16:34:38 +0000 (17:34 +0100)]
mvebu: unify boot.scr creation

Unify boot.scr generation so Makefile for device image generation won't
grow without a reason. Also make boot-scr step optional.

Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
6 years agomvebu: make sdcard bootloader option configurable
Tomasz Maciej Nowak [Wed, 14 Feb 2018 16:28:09 +0000 (17:28 +0100)]
mvebu: make sdcard bootloader option configurable

Remove the necessity for boot loader from SD card image creation process
and make it configurable.

Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
6 years agomvebu: fix partition type and signature for sdcard
Tomasz Maciej Nowak [Wed, 24 Jan 2018 14:32:58 +0000 (15:32 +0100)]
mvebu: fix partition type and signature for sdcard

Previously the partition signature was assigned from provided type. Now
both are corrected wherein signature is always generated from
SOURCE_DATE_EPOCH. With that the root file system can be identified
by PARTUUID string, without relying on static declaration of device node.
This commit also does some cosmetics, removing trailing whitespace and
replacing spaces with tab.

Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
6 years agomvebu: remove redefinition of image name for clearfog
Tomasz Maciej Nowak [Wed, 14 Feb 2018 16:20:00 +0000 (17:20 +0100)]
mvebu: remove redefinition of image name for clearfog

The IMAGE_NAME redefinition causes overwriting of generated SD card
image when multiple root file system types are selected. In result only
single SD card image is generated. This commit fixes this behaviour.

Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
6 years agomvebu: clearfog: use partition uuid for root= bootarg
Josua Mayer [Thu, 19 Oct 2017 07:22:11 +0000 (09:22 +0200)]
mvebu: clearfog: use partition uuid for root= bootarg

U-Boot already knows where it found the boot.scr, and
figuring out the partition UUID becomes trivial at this point.
This change allows booting OpenWrt from whatever storage it has been
flashed to: SD card, eMMC, USB disk or SATA disk.

Signed-off-by: Josua Mayer <josua.mayer97@gmail.com>
[replace lede with openwrt, redact commit message]
Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
6 years agomvebu: Use kernel 4.14 by default
Hauke Mehrtens [Thu, 1 Mar 2018 20:45:24 +0000 (21:45 +0100)]
mvebu: Use kernel 4.14 by default

I am not aware of any regressions in kernel 4.14 compared to kernel 4.9.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agomvebu: activate more workarounds for ARM erratas
Hauke Mehrtens [Sun, 18 Feb 2018 17:57:02 +0000 (18:57 +0100)]
mvebu: activate more workarounds for ARM erratas

The Armada XP uses a Marvell PJ4Bv7 Processor for which already one
workaround for an errata is activated.
The Armada 285 uses a Cortex A9 r4p1 for which the Linux kernel provides
a workaround for ERRATA_764369, activate this.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agomwlwifi: Update to version 10.3.4.0-20180226
Kabuli Chana [Mon, 26 Feb 2018 04:46:55 +0000 (21:46 -0700)]
mwlwifi: Update to version 10.3.4.0-20180226

move to latest version, address vfs_write issue

Signed-off-by: Kabuli Chana <newtownBuild@gmail.com>
6 years agokernel: generic: Move config option to generic
Hauke Mehrtens [Thu, 1 Mar 2018 20:48:28 +0000 (21:48 +0100)]
kernel: generic: Move config option to generic

This is deactivated in all targets using the DSA switch driver with
kernel 4.14.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agogre: squash grev4 and grev6 packages into gre (FS#1399)
Hans Dedecker [Fri, 2 Mar 2018 13:28:43 +0000 (14:28 +0100)]
gre: squash grev4 and grev6 packages into gre (FS#1399)

The split-up into packages gre, grev4 and grev6 causes confusion for the
users as reported in FS#1399.
As IPv4 and IPv6 are considered now as bundled; squash the grev4 and grev6
packages into the gre package and let gre provide both grev4 and grev6.

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agonetifd: add udhcpc link check to dhcp shell handler script
Hans Dedecker [Thu, 1 Mar 2018 17:38:55 +0000 (18:38 +0100)]
netifd: add udhcpc link check to dhcp shell handler script

Fixes the assumption the busybox udhcpc applet is always enabled; in case
the symbolic link check fails the DHCP shell handler script will exit and
as result the DHCP protocol handler will not be registered in netifd.

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agomac80211: brcmfmac: better logging of firmware api errors
Rafał Miłecki [Thu, 1 Mar 2018 17:45:10 +0000 (18:45 +0100)]
mac80211: brcmfmac: better logging of firmware api errors

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: brcmfmac: backport remaining changes from 4.16
Rafał Miłecki [Thu, 1 Mar 2018 17:36:04 +0000 (18:36 +0100)]
mac80211: brcmfmac: backport remaining changes from 4.16

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: brcmfmac: enlarge caps buffer & final sdio restructuring
Rafał Miłecki [Thu, 1 Mar 2018 17:22:51 +0000 (18:22 +0100)]
mac80211: brcmfmac: enlarge caps buffer & final sdio restructuring

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: brcmfmac: restructuring sdio access functions - take 2
Rafał Miłecki [Thu, 1 Mar 2018 17:15:43 +0000 (18:15 +0100)]
mac80211: brcmfmac: restructuring sdio access functions - take 2

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: brcmfmac: restructuring sdio access functions
Rafał Miłecki [Thu, 1 Mar 2018 17:00:19 +0000 (18:00 +0100)]
mac80211: brcmfmac: restructuring sdio access functions

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: brcmfmac: backport remaining changes from 4.15
Rafał Miłecki [Thu, 1 Mar 2018 14:08:25 +0000 (15:08 +0100)]
mac80211: brcmfmac: backport remaining changes from 4.15

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: brcmfmac: firmware halt and scan cleanup
Rafał Miłecki [Thu, 1 Mar 2018 13:55:08 +0000 (14:55 +0100)]
mac80211: brcmfmac: firmware halt and scan cleanup

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: reorder patches putting backports first
Rafał Miłecki [Thu, 1 Mar 2018 13:46:34 +0000 (14:46 +0100)]
mac80211: reorder patches putting backports first

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: add a missing check to the last minstrel fix
Felix Fietkau [Thu, 1 Mar 2018 13:43:56 +0000 (14:43 +0100)]
mac80211: add a missing check to the last minstrel fix

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agomac80211: rename patches accepted upstream
Rafał Miłecki [Thu, 1 Mar 2018 11:32:07 +0000 (12:32 +0100)]
mac80211: rename patches accepted upstream

This is to simplify maintenance. It's easy to say now which patches need
some extra work and/or sending upstream. Updating to newer backports
should be also simpler with this.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agoRevert "ath: do not apply broken power limits with ATH_USER_REGD"
Felix Fietkau [Wed, 28 Feb 2018 20:10:32 +0000 (21:10 +0100)]
Revert "ath: do not apply broken power limits with ATH_USER_REGD"

This reverts commit 79a768a90fa94c0a38c9def1992cc974c091e286.
Some devices can go over their power limits with this commit, so this
needs to be handled on a case by case basis instead

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agomac80211: update the A-MSDU fast-rx patch to the latest version
Felix Fietkau [Wed, 28 Feb 2018 20:14:07 +0000 (21:14 +0100)]
mac80211: update the A-MSDU fast-rx patch to the latest version

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agomac80211: add more minstrel fixes
Felix Fietkau [Thu, 1 Mar 2018 12:34:41 +0000 (13:34 +0100)]
mac80211: add more minstrel fixes

Improves 2.4 GHz throughput

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agomac80211: fix regression introduced by the recent minstrel changes
Felix Fietkau [Thu, 1 Mar 2018 10:59:47 +0000 (11:59 +0100)]
mac80211: fix regression introduced by the recent minstrel changes

The cck rates array needs to be initialized for both the minstrel and
the minstrel_ht variant

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agobase-files: fix off-by-one in counting seconds for factory reset
Rafał Miłecki [Tue, 27 Feb 2018 15:44:25 +0000 (16:44 +0100)]
base-files: fix off-by-one in counting seconds for factory reset

There was a mismatch between indicating factory reset and code actually
starting it. After 5 seconds status LED started blinking rapidly letting
user know it's ready to release reset button. In practice button had to
stay pressed for another second in order to relly start the process.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agoar71xx: fix build with ag71xx debugfs support enabled
Felix Fietkau [Wed, 28 Feb 2018 12:57:58 +0000 (13:57 +0100)]
ar71xx: fix build with ag71xx debugfs support enabled

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agoiw: update to version 4.14
Felix Fietkau [Sat, 24 Feb 2018 10:43:28 +0000 (11:43 +0100)]
iw: update to version 4.14

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agomac80211: add an optimization for fast-rx support
Felix Fietkau [Mon, 26 Feb 2018 21:13:44 +0000 (22:13 +0100)]
mac80211: add an optimization for fast-rx support

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agonetifd: update to the latest version, rewrite RPS/XPS handling
Felix Fietkau [Mon, 26 Feb 2018 11:30:28 +0000 (12:30 +0100)]
netifd: update to the latest version, rewrite RPS/XPS handling

Remove RPS/XPS support from netifd core, move the logic to a hotplug
script that uses a different policy which provides better performance
and more fairness across flows

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agoag71xx: Move timestamp struct member outside of struct.
Rosen Penev [Wed, 7 Feb 2018 21:57:30 +0000 (13:57 -0800)]
ag71xx: Move timestamp struct member outside of struct.

With this change, the timestamp variable is only used in ag71xx_check_dma_stuck. Small tx speedup.

Based on a Qualcomm commit. ag->timestamp = jiffies was not replaced with netif_trans_update(dev) because of this quote:

It should be noted that after this series several instances
of netif_trans_update() are useless (if they occur in
.ndo_start_xmit and driver doesn't set LLTX flag -- stack already
did an update).

From: http://lists.openwall.net/netdev/2016/05/03/87

Signed-off-by: Rosen Penev <rosenp@gmail.com>
6 years agokernel: kmod-geneve: kmod for Geneve tunneling
Yousong Zhou [Tue, 13 Feb 2018 03:20:57 +0000 (11:20 +0800)]
kernel: kmod-geneve: kmod for Geneve tunneling

This will be required for Open vSwitch geneve tunneling support

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
6 years agoramips: fix spurious IRQ7 events when using perf on 4.14
Felix Fietkau [Tue, 27 Feb 2018 14:09:43 +0000 (15:09 +0100)]
ramips: fix spurious IRQ7 events when using perf on 4.14

Upstream handling of MIPS CPU IRQs is rather hackish and the interrupts
are being enabled unconditionally in various places because of legacy
code.
Performance counter events are routed both through the GIC and through
legacy CPU IRQ7 events, causing spurious interrupts.

Fix this by disabling IRQ7 when trying to access the performance counter
IRQ.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agobcm53xx: fix fallback code for picking status LED
Rafał Miłecki [Tue, 27 Feb 2018 15:49:31 +0000 (16:49 +0100)]
bcm53xx: fix fallback code for picking status LED

Looking for a wrong LED file name was stopping this code from find any
LED. This affects devices with only a red/amber power LED.

Fixes: 3aaee1ba023ac ("bcm53xx: failsafe support")
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agonetifd: support DHCP sendopts as list options
Hans Dedecker [Sat, 24 Feb 2018 15:52:38 +0000 (16:52 +0100)]
netifd: support DHCP sendopts as list options

Support config in the form of ....
add_list sendopts=router:10.10.10.2
add_list sendopts=nissrv:20.20.20.2
add_list sendopts=0x7D:abba

This allows to configure sendopts having white spaces as option value

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agoRevert "ramips: improve NAPI polling to increase GRO batch size" (fixes FS#1333)
Felix Fietkau [Tue, 27 Feb 2018 08:41:59 +0000 (09:41 +0100)]
Revert "ramips: improve NAPI polling to increase GRO batch size" (fixes FS#1333)

This reverts commit 968b8616127a52fe5baff20a1f29ff49ade26bc2.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agotoolchain: fix selecting older gcc versions for x86
Felix Fietkau [Tue, 27 Feb 2018 08:00:52 +0000 (09:00 +0100)]
toolchain: fix selecting older gcc versions for x86

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agokernel: m25p80: allow fallback from spi_flash_read to regular SPI transfer
Felix Fietkau [Fri, 23 Feb 2018 16:33:21 +0000 (17:33 +0100)]
kernel: m25p80: allow fallback from spi_flash_read to regular SPI transfer

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agogcc: merge upstream fix for PR 83496 (fixes FS#814)
Felix Fietkau [Sun, 25 Feb 2018 19:25:20 +0000 (20:25 +0100)]
gcc: merge upstream fix for PR 83496 (fixes FS#814)

This bug was causing miscompiled code related to branch delay slots

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agogcc: refresh patches for version 7.3.0
Felix Fietkau [Sun, 25 Feb 2018 19:24:52 +0000 (20:24 +0100)]
gcc: refresh patches for version 7.3.0

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agoiptables: fix compile with kernel 3.18
Hauke Mehrtens [Mon, 26 Feb 2018 16:49:37 +0000 (17:49 +0100)]
iptables: fix compile with kernel 3.18

This fixes a compile bug found by build bot with kernel 3.18

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agokernel: generic: add 4.14 config option
Hans Dedecker [Mon, 26 Feb 2018 15:32:54 +0000 (16:32 +0100)]
kernel: generic: add 4.14 config option

When CGROUPS is enabled the new option CONFIG_CGROUP_NET_CLASSID is
selectable and not handled.
Add this option to the 4.14 kernel configuration.

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agokernel: package efivarfs module
Alif M. Ahmad [Sun, 25 Feb 2018 03:18:41 +0000 (03:18 +0000)]
kernel: package efivarfs module

With this, `mount -t efivarfs` is available and tools such as efitools
and efibootmgr will be usable.

Signed-off-by: Alif M. Ahmad <alive4ever@live.com>
[daniel@makrotopia.org: some whitespace fixes, match From: with SoB]

6 years agosunxi: Add support for Xunlong Orange Pi PC
Vitalij Alshevsky [Sun, 25 Feb 2018 18:12:49 +0000 (21:12 +0300)]
sunxi: Add support for Xunlong Orange Pi PC

CPU: H3 Quad-core Cortex-A7 H.265/HEVC 4K @ 1.3 Ghz
GPU: Mali400MP2 GPU @ 600MHz (supports OpenGL ES 2.0)
Memory: 1GB DDR3 (shared with GPU)
Onboard: Storage TF card (Max. 64GB) / MMC card slot
Onboard: Network 10/100M Ethernet RJ45
Onboard header pinout: SPI, I2C, 1-WIRE
USB 2.0: Three USB 2.0 HOST, One USB 2.0 OTG
Buttons: Power Button(SW4)
Debug TTL UART: ..DC-IN.. >[GND][RX][TX] ..HDMI..

Signed-off-by: Vitalij Alshevsky <v_alshevsky@tut.by>
6 years agouboot-sunxi: add Xunlong Orange Pi PC
Vitalij Alshevsky [Sun, 25 Feb 2018 18:08:57 +0000 (21:08 +0300)]
uboot-sunxi: add Xunlong Orange Pi PC

Signed-off-by: Vitalij Alshevsky <v_alshevsky@tut.by>
6 years agomwlwifi: add and use individual firmware packages
Johnny S. Lee [Tue, 30 Jan 2018 17:19:51 +0000 (01:19 +0800)]
mwlwifi: add and use individual firmware packages

As each mvebu device only uses one of the firmwares provided by mwlwifi
package, it makes sense to put them in separate packages and only install
the one that is needed.

Current mwlwifi version's firmware sizes and usages by devices:
88W8864.bin  118776  caiman, mamba, cobra, shelby
88W8897.bin  489932  (none)
88W8964.bin  449420  rango

Changes by this commit:
 * indicate in title that mwlwifi also is driver for 88W8897 and 88W8964
 * remove mwlwifi package's firmware installation rules
 * add 3 new individual firmware packages (all depends on kmod-mwlwifi):
    - mwlwifi-firmware-88w8864
    - mwlwifi-firmware-88w8897
    - mwlwifi-firmware-88w8964
 * add firmware package to mvebu devices' DEVICE_PACKAGES accordingly

Signed-off-by: Johnny S. Lee <_@jsl.io>
[Add the used FW files to the PACKAGES of default image]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agoscripts/metadata.pm: allow group-only Require-User specs
Matthias Schiffer [Mon, 26 Feb 2018 06:10:32 +0000 (07:10 +0100)]
scripts/metadata.pm: allow group-only Require-User specs

/lib/functions.sh can deal with Require-User specifications that only
contain a group, but no user. Adjust metadata.pm to allow such
specifications as well.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agomvebu: add missing patch for reprobing SFP phys for 4.14
Jonas Gorski [Sun, 25 Feb 2018 21:54:39 +0000 (22:54 +0100)]
mvebu: add missing patch for reprobing SFP phys for 4.14

Add the patch for reprobing phys also for 4.14, as it is still needed.

Fixes: 4ccad922293a ("mvebu: Add support for kernel 4.14")
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
6 years agomvebu: fix SFP insert detection GPIO name on 4.14
Jonas Gorski [Sun, 25 Feb 2018 21:52:41 +0000 (22:52 +0100)]
mvebu: fix SFP insert detection GPIO name on 4.14

When SFP support was accepted upstream, the expected GPIO names were
slightly changed, breaking SFP insert detection. Update the DTS file to
the expected name to make SFP work again.

Fixes: 4ccad922293a ("mvebu: Add support for kernel 4.14")
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
6 years agokernel: more fixes for flow offload
Felix Fietkau [Sun, 25 Feb 2018 17:22:41 +0000 (18:22 +0100)]
kernel: more fixes for flow offload

- fix TCP connection state
- fix checksum error on DNAT

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agoperf: restrict libunwind dependency to archs that actually support libunwind
Matthias Schiffer [Fri, 23 Feb 2018 15:17:57 +0000 (16:17 +0100)]
perf: restrict libunwind dependency to archs that actually support libunwind

Allow building perf on uncommon targets again.

Depending on the kernel version, not all of these archs will actually use
libunwind in perf. Still, it seems simpler and less error-prone to use the
same list that is defined in the libunwind package.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agolibunwind: fix build with musl on PPC
Matthias Schiffer [Fri, 23 Feb 2018 15:17:57 +0000 (16:17 +0100)]
libunwind: fix build with musl on PPC

Works around two incompatiblities between glibc and (POSIX-compliant) musl:

- missing register definitions from asm/ptrace.h
- non-POSIX-compliant ucontext_t on PPC32 with glibc

Compile tested on mpc85xx.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agokernel: fix crash in flow offload code when cleaning up unregistered hooks
Felix Fietkau [Sun, 25 Feb 2018 15:23:07 +0000 (16:23 +0100)]
kernel: fix crash in flow offload code when cleaning up unregistered hooks

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agokernel: remove nf_flow_table hardware offload patch (it is not ready yet)
Felix Fietkau [Sat, 24 Feb 2018 20:04:42 +0000 (21:04 +0100)]
kernel: remove nf_flow_table hardware offload patch (it is not ready yet)

It also does not have any users yet. It will be addde back when the core
API issues have been sorted out

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agokernel: add minimal TCP state tracking to flow offload support
Felix Fietkau [Sun, 25 Feb 2018 14:48:23 +0000 (15:48 +0100)]
kernel: add minimal TCP state tracking to flow offload support

Fixes issues with connections hanging after >30 seconds idle time

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agohostapd: do not register ubus objects for mesh interfaces
Felix Fietkau [Sat, 24 Feb 2018 20:21:09 +0000 (21:21 +0100)]
hostapd: do not register ubus objects for mesh interfaces

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agoodhcp6c: rework sendopts handling
Hans Dedecker [Sat, 24 Feb 2018 19:59:10 +0000 (20:59 +0100)]
odhcp6c: rework sendopts handling

Bring logic of sendopts handling in line with ip6prefix handling

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agoodhcp6c: support multiple additional user prefixes
Jo-Philipp Wich [Wed, 14 Feb 2018 13:58:07 +0000 (14:58 +0100)]
odhcp6c: support multiple additional user prefixes

Support configuration in the form...

    list ip6prefix 2001:db8:1234::/64
    list ip6prefix 2001:db8:5678::/64

... to allow specifying multiple additional IPv6 prefixes.

Implements feature request FS#1361.

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
6 years agoethtool: import from packages, add myself as maintainer
Felix Fietkau [Sat, 24 Feb 2018 15:12:19 +0000 (16:12 +0100)]
ethtool: import from packages, add myself as maintainer

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agoramips: mt7621: add Mediatek AP-MT7621A-V60 EVB support
Roman Yeryomin [Thu, 22 Feb 2018 16:08:14 +0000 (18:08 +0200)]
ramips: mt7621: add Mediatek AP-MT7621A-V60 EVB support

This board has:
- mt7621 SoC
- 8MB SPI flash
- 128MB RAM
- 5x ethernet ports from internal (SoC) switch
- 1x ethernet port sitting on gmac2 and IC+ phy (not yet supported)
- 3x PCIe slots
- 1x USB 2.0 and 1x USB 3.0
- sound based on wm8960
- SDXC card slot (full size)

First fw write from interactive u-boot menu, interrupt with 2.
After that sysupgrade.

Tested both with 4.9 and 4.14

Signed-off-by: Roman Yeryomin <roman@advem.lv>
6 years agoramips: remove chunked-io patch and set spi->max_transfer_size instead
Felix Fietkau [Thu, 22 Feb 2018 09:09:30 +0000 (10:09 +0100)]
ramips: remove chunked-io patch and set spi->max_transfer_size instead

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agokernel: optimize for performance by default starting with 4.14
Felix Fietkau [Thu, 22 Feb 2018 19:18:58 +0000 (20:18 +0100)]
kernel: optimize for performance by default starting with 4.14

Keep size optimizations for smaller targets that already switched

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agokernel: generic: Fix skip GRO for foreign MAC address patches
Hauke Mehrtens [Sat, 24 Feb 2018 11:47:59 +0000 (12:47 +0100)]
kernel: generic: Fix skip GRO for foreign MAC address patches

Somewhere in the kernel 4.9 migration process it looks like this patch
was broken. The update should be done when a device is added to the
bridge and when it gets removed, currently it is only done on the remove
path twice which does not make any sense.
I do not have a setup to test this now so this is only compile tested.

Fixes: f791fb4af450 ("kernel: add linux 4.9 support")
Fixes: b3f95490b9be ("kernel: generic: Add kernel 4.14 support")
Reported-by: Andrey Jr. Melnikov <temnota.am@gmail.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>