openwrt/staging/jogo.git
8 years agoramips: set blocksize for 4MB devices
Mathias Kresin [Sat, 3 Sep 2016 07:56:35 +0000 (09:56 +0200)]
ramips: set blocksize for 4MB devices

Set the blocksize for devices having only 4MB of flash for ramips
devices already using the new image build code.

Informations about the used flash chip are gathered from the OpenWrt
wiki, wikidevi, forums, OEM bootlogs or the compatible property in the
device tree source file.

The en25q32b from the AirLive Air3GII does not have 4k support in the
kernel.

For the following boards no information about the used flash chip could
be found and a 64k blocksize is assumed:

  - Ralink V11ST-FE
  - Ralink AP-RT3052-V22RW-2X2
  - MediaTek MT7628 EVB
  - MediaTek MT7621 EVB
  - UPVEL UR-326N4G
  - Buffalo WZR-AGL300NH

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agoramips: don't set the same max image size twice
Mathias Kresin [Sun, 4 Sep 2016 08:04:20 +0000 (10:04 +0200)]
ramips: don't set the same max image size twice

The ralink_default_fw_size_8M is already set for the default device.

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agoimage: specify max image size in Kilobyte/Megabyte
Mathias Kresin [Sun, 4 Sep 2016 07:58:02 +0000 (09:58 +0200)]
image: specify max image size in Kilobyte/Megabyte

Use the k and m unit suffix to be consistent with the blocksize.

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agoimage: use k as unit suffix for blocksize
Mathias Kresin [Sun, 4 Sep 2016 06:56:05 +0000 (08:56 +0200)]
image: use k as unit suffix for blocksize

Use k as unit suffix for kilobyte to have a the same unit regardless of
the used filesystem.

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agoimage: pass device blocksize to padjffs2
Mathias Kresin [Sat, 3 Sep 2016 05:40:15 +0000 (07:40 +0200)]
image: pass device blocksize to padjffs2

At the moment the padding steps are hardcoded. Especially images for
devices with a 4K sector size can be unnecessarily bloated using the
hardcoded padding steps.

It has been observed that 192Kb of padding was added to the image of a
4MB device, albeit due to the 4K sector size the minimum required extra
padding for the jffs2 rootfs_data is 20Kb.

In worst case it means that the image-size check could fail albeit
there is enough space for all selected packages

For device build code not exposing the blocksize, use the hardcoded
padding further on.

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agokernel: mtd: backport Macronix sector size fix
Mathias Kresin [Sat, 3 Sep 2016 07:27:09 +0000 (09:27 +0200)]
kernel: mtd: backport Macronix sector size fix

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agoapm821xx: image: add support for k unit suffix to boot-img
Mathias Kresin [Thu, 8 Sep 2016 18:14:42 +0000 (20:14 +0200)]
apm821xx: image: add support for k unit suffix to boot-img

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agoimage: add support for k unit suffix to append-ubi
Mathias Kresin [Sun, 4 Sep 2016 06:51:58 +0000 (08:51 +0200)]
image: add support for k unit suffix to append-ubi

Allows to use the same unit for all definitions of the blocksize to be
consistent regardless of the used filesystem.

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agoimage: add support for k unit suffix to pad-offset
Mathias Kresin [Sat, 3 Sep 2016 05:36:12 +0000 (07:36 +0200)]
image: add support for k unit suffix to pad-offset

Allows to specificy the padding and offset in kilobytes to be
consistent with pad-to.

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agoramips: use lower case names for TEW-69xGR images
Mathias Kresin [Fri, 9 Sep 2016 05:12:37 +0000 (07:12 +0200)]
ramips: use lower case names for TEW-69xGR images

Use lower case variant of the device name for to images to be
consistent with all other images names.

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agoramips: fix wrong TARGET_DEVICES
Mathias Kresin [Thu, 8 Sep 2016 21:51:46 +0000 (23:51 +0200)]
ramips: fix wrong TARGET_DEVICES

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agoramips: fix wrong device title
Mathias Kresin [Thu, 8 Sep 2016 05:32:00 +0000 (07:32 +0200)]
ramips: fix wrong device title

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agofirmware-utils: mksenaofw: rework option validation
Mathias Kresin [Wed, 17 Aug 2016 18:08:52 +0000 (20:08 +0200)]
firmware-utils: mksenaofw: rework option validation

The options firmware_type, vendor_id and product_id are not required
for decoding an image file.

Signed-off-by: Mathias Kresin <dev@kresin.me>
8 years agoramips: add support for Mercury MAC1200R v2
Daniel Golle [Thu, 8 Sep 2016 22:57:31 +0000 (00:57 +0200)]
ramips: add support for Mercury MAC1200R v2

MT7628AN + MT7612E, 8MB SPI flash, 64MB DDR RAM
reset button GPIO is still missing (anyone?)
bootloader password is 'slp'

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
8 years agooxnas: small image improvements for kd20
Daniel Golle [Sun, 4 Sep 2016 17:12:12 +0000 (19:12 +0200)]
oxnas: small image improvements for kd20

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
8 years agobcm53xx: enable switch port 5 on Tenda AC9
Rafał Miłecki [Thu, 8 Sep 2016 21:23:41 +0000 (23:23 +0200)]
bcm53xx: enable switch port 5 on Tenda AC9

Now b53 forces port 5 link when needed, we can enable it on Tenda AC9
and start using eth1.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
8 years agokernel: b53: force BCM531x5 port 5 link state if enabled
Rafał Miłecki [Thu, 18 Aug 2016 09:05:12 +0000 (11:05 +0200)]
kernel: b53: force BCM531x5 port 5 link state if enabled

Some devices (e.g. Tenda AC9 based on BCM47189B0) have BCM53125 with
port 5 connected to the second Ethernet interface on the SoC. In such
case there is no PHY and we need to force link manually.

This assumes port 5 can be marked as enabled for such devices. It's not
implemented yet unfortunately.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
8 years agobcm53xx: use upstream queued patches for R8500 and AC9
Rafał Miłecki [Thu, 8 Sep 2016 20:53:00 +0000 (22:53 +0200)]
bcm53xx: use upstream queued patches for R8500 and AC9

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
8 years agokernel: move bcma BCM53573 patch to bcma 4.9 backports
Rafał Miłecki [Thu, 8 Sep 2016 20:39:17 +0000 (22:39 +0200)]
kernel: move bcma BCM53573 patch to bcma 4.9 backports

This patch has been accepted for 4.9.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
8 years agoimx6: override default inittab to add video console tty
Tim Harvey [Tue, 6 Sep 2016 16:44:32 +0000 (09:44 -0700)]
imx6: override default inittab to add video console tty

Adds the following to the default inittab
tty1::askfirst:/bin/ash --login

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
8 years agoimx6: image: remove pca955x/ds1672/at24 from Ventana kernel modules
Tim Harvey [Tue, 6 Sep 2016 16:40:25 +0000 (09:40 -0700)]
imx6: image: remove pca955x/ds1672/at24 from Ventana kernel modules

These need to be static in the kernel for Ventana

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
8 years agoimages: bump default rootfs size to 256 MB
Felix Fietkau [Tue, 6 Sep 2016 15:19:30 +0000 (17:19 +0200)]
images: bump default rootfs size to 256 MB

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agof2fs-tools: import from packages, clean up, and update to latest
Felix Fietkau [Tue, 6 Sep 2016 11:12:17 +0000 (13:12 +0200)]
f2fs-tools: import from packages, clean up, and update to latest

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agodnsmasq: Don't expose *.bind data incl version
Kevin Darbyshire-Bryant [Wed, 7 Sep 2016 10:13:52 +0000 (11:13 +0100)]
dnsmasq: Don't expose *.bind data incl version

Don't expose dnsmasq version & other data to clients via the *.bind
pseudo domain.  This uses a new 'NO_ID' compile time option which has been
discussed and submitted upstream.

This is an alternate to replacing version with 'unknown' which affects
the version reported to syslog and 'dnsmasq --version'

Run time tested with & without NO_ID on Archer C7 v2

Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
8 years agomac80211: remove the fq-disable hack, now that reordering is fixed
Felix Fietkau [Mon, 5 Sep 2016 17:19:48 +0000 (19:19 +0200)]
mac80211: remove the fq-disable hack, now that reordering is fixed

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agomac80211: fix packet loss on fq reordering
Felix Fietkau [Mon, 5 Sep 2016 17:12:05 +0000 (19:12 +0200)]
mac80211: fix packet loss on fq reordering

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agohostapd: update to version 2016-09-05
Felix Fietkau [Tue, 6 Sep 2016 08:53:17 +0000 (10:53 +0200)]
hostapd: update to version 2016-09-05

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agokernel: merge a softirq performance improvement patch
Felix Fietkau [Thu, 1 Sep 2016 10:18:03 +0000 (12:18 +0200)]
kernel: merge a softirq performance improvement patch

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoe2fsprogs: List all libraries explicitly
Florian Fainelli [Wed, 7 Sep 2016 17:25:34 +0000 (10:25 -0700)]
e2fsprogs: List all libraries explicitly

e2fsprogs would fail linking with external toolchains which would not be able
to find several dependencies, explicit them.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
8 years agoe2fsprogs: Honor the global verbose flag
Florian Fainelli [Wed, 7 Sep 2016 17:25:33 +0000 (10:25 -0700)]
e2fsprogs: Honor the global verbose flag

Look for OPENWRT_VERBOSE and pass it down to the e2fsprogs entry-point
Makefile.

Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
8 years agoapm821xx: backport generic HDD led-triggers for WNDR4700 and MBL
Christian Lamparter [Tue, 30 Aug 2016 17:42:39 +0000 (19:42 +0200)]
apm821xx: backport generic HDD led-triggers for WNDR4700 and MBL

This patch backports the new generic HDD/SDD led-trigger
from 4.8-rc1.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
8 years agoapm821xx: detect sd-card media changes for the WNDR4700
Christian Lamparter [Tue, 30 Aug 2016 17:42:38 +0000 (19:42 +0200)]
apm821xx: detect sd-card media changes for the WNDR4700

The insertion or removal of the sd-card cannot be detected
by the hardware itself. This is by design. To workaround this,
for the WNDR4700 unload/load the dwc2 module in case the
the special SD CARD GPIO line is low/high.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
8 years agocns3xxx: add GW2386 support
Tim Harvey [Wed, 7 Sep 2016 21:05:42 +0000 (14:05 -0700)]
cns3xxx: add GW2386 support

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
8 years agocns3xxx: add GW2394 Support
Tim Harvey [Wed, 7 Sep 2016 15:47:41 +0000 (08:47 -0700)]
cns3xxx: add GW2394 Support

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
8 years agoimage.mk: Create a manifest file of installed packages as a build artifact
Pushpal Sidhu [Thu, 1 Sep 2016 14:54:03 +0000 (07:54 -0700)]
image.mk: Create a manifest file of installed packages as a build artifact

A few linux BSP's create a manifest file of installed packages for a given
target in order to help them understand exactly what's on their images. Create
one here as well as a build artifact since many users have an affinity to
prune down on packages to save valuable flash space.

Signed-off-by: Pushpal Sidhu <psidhu@gateworks.com>
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
8 years agokernel: backport upstream overlayfs fixes
Rafał Miłecki [Tue, 6 Sep 2016 19:32:47 +0000 (21:32 +0200)]
kernel: backport upstream overlayfs fixes

First two patches weren't marked for stable but are dependencies for
laters ones. The rest of patches was marked for stable but most likely
will be backported to 4.5+ only so we need to get them on our own.

An important fix is eea2fb4851e9d ("ovl: proper cleanup of workdir") as
it allows mounting overlayfs with dirty workdir, e.g. after power cut.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
8 years agotarget/toolchain: Fix toolchain packaging without package build
Micha Lenk [Tue, 9 Aug 2016 10:59:06 +0000 (12:59 +0200)]
target/toolchain: Fix toolchain packaging without package build

If the toolchain is packaged for later use as external toolchain, the resulting
tarball is created in $BIN_DIR. But without building all packages first that
directory isn't created, hence 'make target/toolchain/compile' fails when
trying to create the toolchain tarball with error "Cannot open: No such file or
directory".

To fix that the $BIN_DIR is created before using it.

Signed-off-by: Micha Lenk <micha@lenk.info>
8 years agodnsmasq: fix remove pidfile on shutdown regression
Kevin Darbyshire-Bryant [Mon, 5 Sep 2016 08:01:19 +0000 (09:01 +0100)]
dnsmasq: fix remove pidfile on shutdown regression

Regression introduced by 3481d0d dnsmasq: run as dedicated UID/GID

dnsmasq is unable to remove its own pidfile as /var/run/dnsmasq is owned
by root and now dnsmasq runs as dnsmasq:dnsmasq.  Change directory
ownership to match.

dnsmasq initially starts as root, creates the pidfile, then drops to
requested non-root user.  Until this fix dnsmasq had insufficient
privilege to remove its own pidfile.

Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
8 years agomwlwifi: Expose the IEEE 802.11w support to hostapd
Petko Bordjukov [Mon, 5 Sep 2016 15:27:13 +0000 (18:27 +0300)]
mwlwifi: Expose the IEEE 802.11w support to hostapd

Add a dependency on DRIVER_11W_SUPPORT in order to enable the IEEE
802.11w functionality in hostapd.

Signed-off-by: Petko Bordjukov <bordjukov@gmail.com>
8 years agohostapd: fix typo and indentation in ap_sta_support.patch
Johannes Römer [Sun, 4 Sep 2016 11:23:24 +0000 (13:23 +0200)]
hostapd: fix typo and indentation in ap_sta_support.patch

Signed-off-by: Johannes Römer <jroemer@posteo.net>
8 years agoath10k-ct: fix missing symbols if ath9k is not selected
Alexis Green [Fri, 2 Sep 2016 18:58:07 +0000 (11:58 -0700)]
ath10k-ct: fix missing symbols if ath9k is not selected

Require kernel to compile with CONFIG_RELAY=y

Signed-off-by: Alexis Green <alexis@cessp.it>
8 years agomac80211: backport new register bitfield macros
Felix Fietkau [Sun, 4 Sep 2016 15:39:12 +0000 (17:39 +0200)]
mac80211: backport new register bitfield macros

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agomac80211: add a tx sequence number allocation fix
Felix Fietkau [Sun, 4 Sep 2016 15:58:08 +0000 (17:58 +0200)]
mac80211: add a tx sequence number allocation fix

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoapm821xx: add size check for initramfs kernel for the Meraki MR24
Christian Lamparter [Fri, 2 Sep 2016 20:35:03 +0000 (22:35 +0200)]
apm821xx: add size check for initramfs kernel for the Meraki MR24

Russell Senior reported an issue with the MR24 initramfs kernels:
> ## Booting kernel from Legacy Image at 00c10000 ...
>   Image Name:   POWERPC LEDE Linux-4.4.19
>   Created:      2016-08-31  11:57:05 UTC
>   Image Type:   PowerPC Linux Kernel Image (gzip compressed)
>   Data Size:    2155723 Bytes = 2.1 MiB
>   Load Address: 00000000
>   Entry Point:  00000000
>   Verifying Checksum ... OK
>Wrong Ramdisk Image Format
>Ramdisk image is corrupt or invalid

For the MR24, the kernel is uploaded to 0x10000. The ramdisk starts
at 0x200000. This leaves the kernel with just 0x1f0000 bytes = 1984kb.
This patch adds a size check so the image creation script will abort
instead of producing a unbootable initramfs image. A separate patch
"apm821xx: Fix initramfs image for the Meraki MR24" which fixed the
reported issue was submitted earlier.

Cc: Russell Senior <russell@personaltelco.net>
Cc: Chris Blake <chrisrblake93@gmail.com>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
8 years agoapm821xx: Fix initramfs image for the Meraki MR24
riptidewave93 [Fri, 2 Sep 2016 11:17:24 +0000 (06:17 -0500)]
apm821xx: Fix initramfs image for the Meraki MR24

When gzip is used, our kernel is too large and this causes the ramdisk
to be at the wrong offset. Fix by moving to lzma.

Signed-off-by: Chris Blake <chrisrblake93@gmail.com>
8 years agodropbear: mdns flag is a bool, not integer
Karl Palsson [Thu, 25 Aug 2016 11:48:30 +0000 (11:48 +0000)]
dropbear: mdns flag is a bool, not integer

Effectively the same for most purposes, but more accurate.

Signed-off-by: Karl Palsson <karlp@etactica.com>
8 years agobase-files: support oneshot leds properly.
Karl Palsson [Fri, 26 Aug 2016 11:17:49 +0000 (11:17 +0000)]
base-files: support oneshot leds properly.

oneshot trigger configurations for LEDs are created, but the on/off
timing configurations are ignored.  generate_config is correctly creating
oneshot configs, but the later led script doesn't recognise the trigger
details.

Fixes: c0c3f2d4c917 leds: support oneshot as well as timer triggers
Signed-off-by: Karl Palsson <karlp@etactica.com>
8 years agofstools: fix logic bug in extroot verification code
John Crispin [Mon, 5 Sep 2016 05:09:27 +0000 (07:09 +0200)]
fstools: fix logic bug in extroot verification code

Signed-off-by: John Crispin <john@phrozen.org>
8 years agougps: update to latest git HEAD
John Crispin [Wed, 24 Aug 2016 12:10:22 +0000 (14:10 +0200)]
ugps: update to latest git HEAD

Signed-off-by: John Crispin <john@phrozen.org>
8 years agozram-swap: CONFIG_PROCD_ZRAM_TMPFS compatibility
Conn O'Griofa [Sun, 24 Jul 2016 16:18:39 +0000 (17:18 +0100)]
zram-swap: CONFIG_PROCD_ZRAM_TMPFS compatibility

Enable CONFIG_PROCD_ZRAM_TMPFS compatibility via two changes to list_cpu_idx():
* detect if /tmp is being used by /dev/zram0; if yes, offset initial value by 1 to skip first zram device.
* hot-add /dev/zram1, if not already present.

Signed-off-by: Conn O'Griofa >connogriofa@gmail.com>
8 years agoimx6: enable ARM crypto acceleration
Koen Vandeputte [Thu, 1 Sep 2016 14:10:01 +0000 (16:10 +0200)]
imx6: enable ARM crypto acceleration

8 years agoimx6: refresh kernel config
Koen Vandeputte [Thu, 1 Sep 2016 14:07:00 +0000 (16:07 +0200)]
imx6: refresh kernel config

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
8 years agogeneric: add NET3380 UDC support
Tim Harvey [Thu, 1 Sep 2016 14:55:15 +0000 (07:55 -0700)]
generic: add NET3380 UDC support

Add a patch to backport 5185c91385d73cdf79836eb8548e4726e43ae831
from Linux 4.8 adding USB2380 support to the NET2280 driver and
create an OpenWrt menu option to select it as a module.

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
8 years agoimx6: add GW553x support
Tim Harvey [Thu, 1 Sep 2016 14:49:28 +0000 (07:49 -0700)]
imx6: add GW553x support

The Gateworks Ventana GW553x is a single-board computer based on the NXP
IMX6 SoC with the following features:
 * IMX6 DualLite Soc (supports IMX6S,IMX6DL,IMX6Q)
 * small form factor (35x70mm)
 * 512MB DDR3 DRAM (2x32bit) (options up to 2GB)
 * 256MB NAND flash (4GB option)
 * Gateworks System Controller:
  - hardware watchdog
  - hardware monitor
  - pushbutton controller
  - EEPROM storage
  - power control
 * JTAG programmable
 * 1x miniPCI socket (with PCIe, USB, SIM)
 * 1x microSD socket
 * 1x SIM socket
 * Inertial Module (LSM9DS1 9DOF: 3x acc, 3x rate, 3x mag)
 * analog CVBS video in
 * GPS (optional uBlox EVA-M8M)
 * Application headers:
  - 2x TTL UART (TX/RX)
  - 4x TTL GPIO (3x configurable as PWM)
  - 3x CVBS Analog video input (1x decoder with 3x selectable inputs)
 * Front panel connectors:
  * micro-HDMI audio/video out
  - 1x user programmable LED
  - 1x configurable user pushbutton
  - 1x USB OTG

See http://www.gateworks.com for more info

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
8 years agoboot: kobs-ng: update kobs-ng for newer kernels
Tim Harvey [Thu, 1 Sep 2016 14:45:37 +0000 (07:45 -0700)]
boot: kobs-ng: update kobs-ng for newer kernels

This allows kobs-ng to flash the SPL successfully on the 4.4 kernel used by
the Gateworks IMX boards supporting NAND. The previous version of kobs-ng
worked with the 3.14 kernel but will brick a board making its SPL unbootable
for the 4.4 kernel.

See http://trac.gateworks.com/wiki/ventana/bootloader#nandspl for instructions
on updating the SPL from Linux.

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
8 years agotoolchain/gcc: bump GCC 6.1.0 to 6.2.0
Syrone Wong [Sat, 3 Sep 2016 10:31:14 +0000 (18:31 +0800)]
toolchain/gcc: bump GCC 6.1.0 to 6.2.0

Compile and run tested on ar71xx and mvebu

Signed-off-by: Syrone Wong <wong.syrone@gmail.com>
8 years agobuild: fix subtarget descriptions
Felix Fietkau [Sun, 4 Sep 2016 11:34:18 +0000 (13:34 +0200)]
build: fix subtarget descriptions

Move Target/Description above the target.mk include

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agooxnas: add mem=128M to stg-212 built-in cmdline for legacy u-boot
Daniel Golle [Sun, 4 Sep 2016 02:56:31 +0000 (04:56 +0200)]
oxnas: add mem=128M to stg-212 built-in cmdline for legacy u-boot

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
8 years agooxnas: add method to extract mac_adr from legacy cmdline
Daniel Golle [Sun, 4 Sep 2016 01:34:19 +0000 (03:34 +0200)]
oxnas: add method to extract mac_adr from legacy cmdline

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
8 years agooxnas: kd20: generate image compatible with stock firmware
Daniel Golle [Sun, 4 Sep 2016 01:11:45 +0000 (03:11 +0200)]
oxnas: kd20: generate image compatible with stock firmware

 * adjust MTD partitions to match vendor's scheme
 * generate *factory.tar.gz image compatible with stock web-if
   firmware upgrade interface

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
8 years agooxnas: allow booting with legacy loaders
Daniel Golle [Sun, 4 Sep 2016 00:09:01 +0000 (02:09 +0200)]
oxnas: allow booting with legacy loaders

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
8 years agoar71xx: fix ethernet driver fast reset issue causing tx timeouts
Felix Fietkau [Sat, 3 Sep 2016 11:46:18 +0000 (13:46 +0200)]
ar71xx: fix ethernet driver fast reset issue causing tx timeouts

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoramips: fix legacy initramfs images
Felix Fietkau [Sat, 3 Sep 2016 08:40:11 +0000 (10:40 +0200)]
ramips: fix legacy initramfs images

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoath9k: fix regression in tx queueing patch
Felix Fietkau [Fri, 2 Sep 2016 10:10:51 +0000 (12:10 +0200)]
ath9k: fix regression in tx queueing patch

power save response frames can go through the old tx path, and the tid
needs to be set for sequence numbers to be assigned correctly.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agomvebu: use ptgen instead of fdisk to make the clearfog image script more portable
Felix Fietkau [Thu, 1 Sep 2016 13:29:48 +0000 (15:29 +0200)]
mvebu: use ptgen instead of fdisk to make the clearfog image script more portable

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agomvebu: fix boot script for booting from mmc
Felix Fietkau [Thu, 1 Sep 2016 13:29:14 +0000 (15:29 +0200)]
mvebu: fix boot script for booting from mmc

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agomvebu: add sdcard image creation script
Josua Mayer [Tue, 9 Aug 2016 13:23:24 +0000 (15:23 +0200)]
mvebu: add sdcard image creation script

Added gen_mvebu_sdcard_img.sh to facilitate creating an fixed-size sdcard image,
adding the bootloader and populating it with actual data.

Added the required rules for creating a 4GB sdcard image according to this layout:
p0: boot (fat32)
p1: rootfs (squashfs)
p2: rootfs_data (ext4)
This should be generic to any mvebu boards that can boot from block storage.

Added the new sdcard image to the Clearfog image profile.

Signed-off-by: Josua Mayer <josua.mayer97@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name> [cleanup]
8 years agomvebu: clearfog: require uboot-mvebu-clearfog
Josua Mayer [Sat, 13 Aug 2016 13:50:54 +0000 (15:50 +0200)]
mvebu: clearfog: require uboot-mvebu-clearfog

The u-boot variant for the clearfog is provided by the uboot-mvebu-clearfog,
and not by the generic uboot-mvebu packae.
Make sure the clearfog variant is selected when building for it.

Signed-off-by: Josua Mayer <josua.mayer97@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoath10k-firmware: add QCA9887 firmware
Zhao Gang [Mon, 22 Aug 2016 10:03:36 +0000 (10:03 +0000)]
ath10k-firmware: add QCA9887 firmware

QCA9887 is experimentally supported in compat-wireless-2016-06-20.

Signed-off-by: Zhao Gang <gang.zhao.42@gmail.com>
8 years agoAdd config symbols for kernel keyring support
Nathaniel Wesley Filardo [Sun, 19 Oct 2014 07:28:13 +0000 (03:28 -0400)]
Add config symbols for kernel keyring support

Enable selection of the kernel key retention framework and some of its
additional facilities; see Documentation/security/keys.txt and
security/keys/Kconfig for details

Signed-off-by: Nathaniel Wesley Filardo <nwfilardo@gmail.com>
8 years agomwlwifi: Updated to latest source
Gabe Rodriguez [Wed, 31 Aug 2016 18:16:49 +0000 (11:16 -0700)]
mwlwifi: Updated to latest source

This commit updates the mwlwifi driver to the latest version provided in the repo.

Signed-off-by: Gabe Rodriguez <lifehacksback@gmail.com>
8 years agomac80211: add fixes for dealing with unexpected BlockAck frames
Felix Fietkau [Tue, 30 Aug 2016 11:12:45 +0000 (13:12 +0200)]
mac80211: add fixes for dealing with unexpected BlockAck frames

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoath9k: add a bunch of powersave handling fixes
Felix Fietkau [Sun, 28 Aug 2016 11:50:09 +0000 (13:50 +0200)]
ath9k: add a bunch of powersave handling fixes

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agomac80211: add a powersave handling fix
Felix Fietkau [Sun, 28 Aug 2016 11:49:48 +0000 (13:49 +0200)]
mac80211: add a powersave handling fix

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoAdd Issue submission template (redirect to bugs.lede-project.org)
Ted Hess [Tue, 30 Aug 2016 19:27:46 +0000 (15:27 -0400)]
Add Issue submission template (redirect to bugs.lede-project.org)

Signed-off-by: Ted Hess <thess@kitschensync.net>
8 years agoopenssl: re-enable ARM assembly
Felix Fietkau [Wed, 31 Aug 2016 11:31:38 +0000 (13:31 +0200)]
openssl: re-enable ARM assembly

The original reason for disabling it seems to have been fixed
Related discussion: https://github.com/lede-project/source/pull/307

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoimx6: disable EOF markers on UBI
Felix Fietkau [Wed, 31 Aug 2016 11:06:54 +0000 (13:06 +0200)]
imx6: disable EOF markers on UBI

The boot loader does not recognize them and might mess up the volume

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoscripts/ubinize-image.sh: add support for adding custom partitions
Felix Fietkau [Sun, 31 Jul 2016 16:34:02 +0000 (18:34 +0200)]
scripts/ubinize-image.sh: add support for adding custom partitions

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoglibc: re-enable parallel builds
Felix Fietkau [Tue, 30 Aug 2016 09:12:39 +0000 (11:12 +0200)]
glibc: re-enable parallel builds

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoglibc: switch to 2.24 by default and remove old versions, fixes security issues
Felix Fietkau [Tue, 30 Aug 2016 09:02:54 +0000 (11:02 +0200)]
glibc: switch to 2.24 by default and remove old versions, fixes security issues

2.24 fixes the following CVEs compared to 2.23:
- CVE-2016-3075
- CVE-2016-3706
- CVE-2016-1234
- CVE-2016-4429
- CVE-2016-5417

CVEs fixed in 2.23:
- CVE-2015-8776
- CVE-2015-8777
- CVE-2015-8778
- CVE-2015-8779
- CVE-2014-9761
- CVE-2015-7547

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoebtables: fix build with glibc
Felix Fietkau [Tue, 30 Aug 2016 10:04:01 +0000 (12:04 +0200)]
ebtables: fix build with glibc

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agodante: remove -D_GNU_SOURCE to fix build errors with current glibc
Felix Fietkau [Tue, 30 Aug 2016 09:49:46 +0000 (11:49 +0200)]
dante: remove -D_GNU_SOURCE to fix build errors with current glibc

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoglibc: add 2.24
Dirk Neukirchen [Mon, 29 Aug 2016 12:16:14 +0000 (14:16 +0200)]
glibc: add 2.24

since eglibc is deprecated just use glibc as name

- fix build errors w. stack-protector (observed for x86_64):
libresolv_pic.a(gethnamaddr.os): In function `addrsort':
gethnamaddr.c:(.text+0x425): undefined reference to `__stack_chk_guard'

by additionally setting libc_cv_ssp_strong=no
like previously libc_cv_ssp=no was set

- fix compile errors on 2.24 x86 introduced by
upstream changesets "Optimize i386 syscall inlining"

errors are like:
- https://sourceware.org/ml/libc-alpha/2015-10/msg00745.html
and trying to fix with the whats suggested at:
https://patchwork.openembedded.org/patch/118909/
leads to other error:
gcc6: elf/librtld.os: In function `__mmap':
(.text+0x131a9): undefined reference to `__libc_do_syscall'
or:
gcc5: elf/dl-load.os
{standard input}: Assembler messages:
{standard input}:5129: Error: symbol `__x86.get_pc_thunk.cx' is already defined

instead of testing other flags/effects (-fno-omit-frame-pointer)
just use -O2 (like buildroot does) instead of -Os

boot+pings tested on:
qemu malta (le+be) (gcc5+bin2.25.1)
qemu aarch64 (gcc5+bin2.25.1)
qemu x86_64 (gcc6+bin2.26)
qemu x86 generic (gcc6+bin2.26)

Signed-off-by: Dirk Neukirchen <dirkneukirchen@web.de>
8 years agoiperf: add -lm to fix build with newer glibc
Felix Fietkau [Mon, 29 Aug 2016 13:09:50 +0000 (15:09 +0200)]
iperf: add -lm to fix build with newer glibc

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoiperf: drop PKG_BUILD_DIR override
Felix Fietkau [Mon, 29 Aug 2016 13:09:03 +0000 (15:09 +0200)]
iperf: drop PKG_BUILD_DIR override

No longer necessary since the removal of build variants

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agox86: enable fpu for all subtargets
Felix Fietkau [Mon, 29 Aug 2016 12:39:20 +0000 (14:39 +0200)]
x86: enable fpu for all subtargets

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agomvebu: add fpu to features
Josua Mayer [Tue, 16 Aug 2016 14:02:43 +0000 (16:02 +0200)]
mvebu: add fpu to features

This triggers HAS_FPU=y, and unsets CONFIG_SOFT_FLOAT.
Considering all mvebu boards have an fpu, this is the desirable behaviour.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agomvebu: enable ARM crypto acceleration
Felix Fietkau [Mon, 29 Aug 2016 11:51:58 +0000 (13:51 +0200)]
mvebu: enable ARM crypto acceleration

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agomvebu: enable NEON support
Felix Fietkau [Mon, 29 Aug 2016 11:51:27 +0000 (13:51 +0200)]
mvebu: enable NEON support

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agomvebu: make MMC driver for Clearfog builtin
Josua Mayer [Mon, 15 Aug 2016 14:20:45 +0000 (16:20 +0200)]
mvebu: make MMC driver for Clearfog builtin

microSD is the primary boot media for the Clearfog Pro board, and should
work unconditionally.

Signed-off-by: Josua Mayer <josua.mayer97@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name> [cleanup]
8 years agomvebu: refresh kernel config
Felix Fietkau [Mon, 29 Aug 2016 11:41:16 +0000 (13:41 +0200)]
mvebu: refresh kernel config

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agokernel: spi-nor: add support for ESMT_f25l32qa and ESMT_f25l64qa
Felix Fietkau [Sun, 28 Aug 2016 13:31:55 +0000 (15:31 +0200)]
kernel: spi-nor: add support for ESMT_f25l32qa and ESMT_f25l64qa

Add Support for the ESMT_F25L32QA and ESMT_F25L64QA.
These are 4MB and 8MB SPI NOR Chips from Elite Semiconductor Memory Technology

Signed-off-by: L. D. Pinney <ldpinney@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agokernel: make serial port sysrq-disable patch more generic (FS#112)
Felix Fietkau [Sun, 28 Aug 2016 12:43:12 +0000 (14:43 +0200)]
kernel: make serial port sysrq-disable patch more generic (FS#112)

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoRevert "opkg: use vfork on gz_open by default (FS#120)"
Jo-Philipp Wich [Sun, 28 Aug 2016 21:01:02 +0000 (23:01 +0200)]
Revert "opkg: use vfork on gz_open by default (FS#120)"

This reverts commit 763f5d787316ec1b041450a644ecd9416f11e94f.

Currently the vfork() code path in opkg is broken and relies on unsupported
ftello() / fseeko() operations on pipes - we need to restructure the code
before we can reconsider this approach.

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
8 years agoRevert "opkg: disable the use of vfork for the host build"
Jo-Philipp Wich [Sun, 28 Aug 2016 20:56:16 +0000 (22:56 +0200)]
Revert "opkg: disable the use of vfork for the host build"

This reverts commit 02e3c718e9ad0c14aa3cf359ba2a4763e81df805.

Currently the vfork() code path in opkg is broken and relies on unsupported
ftello() / fseeko() operations on pipes - we need to restructure the code
before we can reconsider this approach.

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
8 years agoath10k-firmware: Update to latest ath10k-ct 9984 firmware.
Ben Greear [Thu, 25 Aug 2016 22:10:29 +0000 (15:10 -0700)]
ath10k-firmware: Update to latest ath10k-ct 9984 firmware.

Tested briefly on Netgear r7800.  Firmware failed to load on first
boot, but then it worked after that and I could not reproduce the
failure.

Signed-off-by: Ben Greear <greearb@candelatech.com>
8 years agotools: make mtools/dosfstools unconditional
Felix Fietkau [Sun, 28 Aug 2016 18:58:12 +0000 (20:58 +0200)]
tools: make mtools/dosfstools unconditional

The list of targets that need them are getting longer

Signed-off-by: Felix Fietkau <nbd@nbd.name>
8 years agoar71xx: generate US- and EU-specific images for the Archer C7 v2
Matthias Schiffer [Sun, 28 Aug 2016 17:49:48 +0000 (19:49 +0200)]
ar71xx: generate US- and EU-specific images for the Archer C7 v2

TP-Link has started providing US- and EU-specific stock firmwares that only
allow upgrading to firmwares with the same region code. Provide factory
images for both these regions.

To avoid confusing users outside these regions, we still provide a
"universal" factory.bin without a region code, although flashing either of
the US and EU images would work as well.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
8 years agofirmware-utils: mktplinkfw: add support for TP-Link's new region codes
Matthias Schiffer [Sun, 28 Aug 2016 17:43:25 +0000 (19:43 +0200)]
firmware-utils: mktplinkfw: add support for TP-Link's new region codes

TP-Link has changed the way the region is stored in the firmware header,
and now provides US- and EU-specific images for the Archer C7. Adding the
new region codes is necessary to make LEDE/OpenWrt flashable on devices
with the new stock firmwares again.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
8 years agoopkg: disable the use of vfork for the host build
Felix Fietkau [Sun, 28 Aug 2016 12:33:48 +0000 (14:33 +0200)]
opkg: disable the use of vfork for the host build

Signed-off-by: Felix Fietkau <nbd@nbd.name>