Rafał Miłecki [Tue, 13 Sep 2016 09:28:07 +0000 (11:28 +0200)]
mac80211: stop brcmfmac from selecting all PCIe firmwares
Now we have firmwares separated and bcm53xx selecting required ones make
use of it to actually save that rootfs space.
Other targets using brcmfmac (brcm2708 and sunxi) use SDIO interface and
firmware so they don't won't be affected.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
John Crispin [Wed, 14 Sep 2016 19:37:21 +0000 (21:37 +0200)]
mountd: update to latest git HEAD
fixes cleanup of mount points
Signed-off-by: John Crispin <john@phrozen.org>
Mathias Kresin [Tue, 13 Sep 2016 19:47:25 +0000 (21:47 +0200)]
ramips: enable 4K sector support in kernel config
This is a follow up to
28110727f1d513a6c4c13aa87ecb606aa5d306eb
"ramips: set blocksize for 4MB devices". I've missed to include the
required changes of the kernel configs to enable 4K sector size
support.
The option is only enabled for targets having boards with 4k sector
size flash chips.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Sun, 11 Sep 2016 17:05:20 +0000 (19:05 +0200)]
image: remove padding parameter from append-kernel/append-rootfs
Using pad-to instead of passing the optional padding to append-kernel
or append-rootfs. It could be that the value of a variable is passed.
In case the variable is empty no error is thrown.
Furthermore the purpose of the extra parameter is hard to get without
reading the code.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Tue, 13 Sep 2016 05:18:10 +0000 (07:18 +0200)]
image: use check-size from new image build code
Use the new image build code and deduplicate the code.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Tue, 13 Sep 2016 17:23:52 +0000 (19:23 +0200)]
image: add KERNEL_SIZE to the default device vars
This reverts commit
ec37a565878fd79d4127cbb843c17ce3b8dc97c0 and fixes
the underlying issue.
Rafał Miłecki [Tue, 13 Sep 2016 10:24:36 +0000 (12:24 +0200)]
bcm53xx: use the latest submitted version of ILP clock driver
It switched to syscon to better describe PMU block.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Rafał Miłecki [Tue, 13 Sep 2016 10:05:33 +0000 (12:05 +0200)]
hostapd: fix regression breaking brcmfmac
The latest update of hostapd broke brcmfmac due to upstream regression.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Rafał Miłecki [Tue, 13 Sep 2016 09:03:02 +0000 (11:03 +0200)]
bcm53xx: specify brcmfmac firmware for every device
This will allow us to include required firmwares only and save some
rootfs space.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Rafał Miłecki [Sun, 11 Sep 2016 20:56:45 +0000 (22:56 +0200)]
linux-firmware: separate packages for Broadcom FullMAC PCIe firmwares
Using few packages will allow saving some space by decreasing rootfs
size. Dropping
43602a1 firmware saves 316 580 B. Dropping 4366b1 saves
468 188 B.
Moreover there are more firmware files that may require packaging and
even more to come later (e.g. 4366c0).
This can especially useful now, with per device rootfs.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Daniel Gimpelevich [Sun, 17 Jul 2016 15:30:42 +0000 (08:30 -0700)]
kernel: Backport pending appended DTB handling patches
Backport patches from upstream Linux kernel which are making the
kernel stores the appended dtb not in the same resisters as defined in
the UHI specification, use a separate variable on MIPS.
Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
[some modifications]
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Felix Fietkau [Mon, 12 Sep 2016 16:20:50 +0000 (18:20 +0200)]
build: drop UBI EOF marker from images by default
Only add them where they are actually required.
Should help with compatibility issues with stock U-Boot images that
access UBI
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Mon, 12 Sep 2016 09:25:20 +0000 (11:25 +0200)]
mac80211: fix tx issue with CCMP PN generated in hardware
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Mathias Kresin [Sun, 11 Sep 2016 15:46:09 +0000 (17:46 +0200)]
ramips: improve Linksys EA8500 build code
Use of the kernel size variable.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Sun, 11 Sep 2016 17:06:32 +0000 (19:06 +0200)]
mvebu: fix OpenBlocks AX3 image
The KERNEL_SIZE variable is unset and no padding is applied. This looks
like a typo to me since the ubinized image need to be aligned to the
flash blocksize.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Sun, 11 Sep 2016 15:50:29 +0000 (17:50 +0200)]
ar71xx: fix typo in image build code
The kernel size isn't passed to the append-kernel build step and the
resulting image bricks the device.
Fixes FS#168.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Sat, 10 Sep 2016 07:46:14 +0000 (09:46 +0200)]
cleanup ucidef_set_interface* usage
- quote the interface name
- remove call of not existing function
- remove the proto if it's the default proto
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Fri, 9 Sep 2016 19:10:54 +0000 (21:10 +0200)]
use immediate set in target Makefiles
No functional change, it's just to have the same style everywhere. This
way I don't need to use any regex magic to extract all subtargets for
compile tests.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Fri, 9 Sep 2016 18:57:58 +0000 (20:57 +0200)]
ramips: board.d merge redundant switch/interface configs
Fix alphabetical order distortions as well.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Thu, 25 Aug 2016 20:19:00 +0000 (22:19 +0200)]
lantiq: board.d: apply alphabetical order
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Tue, 23 Aug 2016 05:36:00 +0000 (07:36 +0200)]
lantiq: board.d: cleanup default switch config
Add or fix the physical port number of switch ports to allow LuCI to
display the ports in physical order.
Remove the physical port number in case bridge port and physical port
order are the same.
Reorder the ucidef_add_switch parameter to be aligned with the switch
port number.
Remove board settings which are covered by the default case and remove
comments.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Thu, 18 Aug 2016 19:14:41 +0000 (21:14 +0200)]
lantiq: enable cpu temp driver for more tested boards
Signed-off-by: Mathias Kresin <dev@kresin.me>
Álvaro Fernández Rojas [Sat, 10 Sep 2016 12:54:26 +0000 (14:54 +0200)]
brcm2708: update linux 4.4 patches to latest version
As usual these patches were extracted and rebased from the raspberry pi repo:
https://github.com/raspberrypi/linux/tree/rpi-4.4.y
Also adds support for Raspberry Pi Compute Module 3 (untested).
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Álvaro Fernández Rojas [Sat, 10 Sep 2016 12:54:08 +0000 (14:54 +0200)]
brcm2708-gpu-fw: update to latest version
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Kevin Darbyshire-Bryant [Fri, 9 Sep 2016 08:03:27 +0000 (09:03 +0100)]
dnsmasq: make NO_ID optional in full variant
Permit users of the full variant to disable the NO_ID *.bind pseudo
domain masking.
Defaulted 'on' in all variants.
Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
Kevin Darbyshire-Bryant [Wed, 7 Sep 2016 16:23:51 +0000 (17:23 +0100)]
dropbear: hide dropbear version
As security precaution and to limit the attack surface based on
the version reported by tools like nmap mask out the dropbear
version so the version is not visible anymore by snooping on the
wire. Version is still visible by 'dropbear -V'
Based on a patch by Hans Dedecker <dedeckeh@gmail.com>
Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
Signed-off-by: Felix Fietkau <nbd@nbd.name> [remove trailing _]
Felix Fietkau [Thu, 8 Sep 2016 12:48:43 +0000 (14:48 +0200)]
x86: disable MTD support
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 13:16:12 +0000 (15:16 +0200)]
octeon: remove block2mtd support
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 12:46:49 +0000 (14:46 +0200)]
octeon: use new ext4/f2fs overlay support
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 12:46:37 +0000 (14:46 +0200)]
octeon: enable f2fs and loopback device support in the config
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 12:46:25 +0000 (14:46 +0200)]
octeon: enable f2fs and ext4 utilities
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 12:19:34 +0000 (14:19 +0200)]
mvebu: use ext4/f2fs overlay filesystem for solidrun clearfog
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 11:59:32 +0000 (13:59 +0200)]
mvebu: enable ext4, f2fs and loop device support
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 11:40:12 +0000 (13:40 +0200)]
x86: drop the use of block2mtd, use ext4/f2fs as overlay filesystem
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 11:39:14 +0000 (13:39 +0200)]
x86: enable mkf2fs and e2fsprogs by default
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Tue, 6 Sep 2016 11:46:44 +0000 (13:46 +0200)]
x86: enable f2fs and loopback device support in the kernel config
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Felix Fietkau [Thu, 8 Sep 2016 11:41:44 +0000 (13:41 +0200)]
fstools: update to the latest version, adds support for ext4/f2fs overlay via loopback device
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Koen Vandeputte [Fri, 9 Sep 2016 07:44:55 +0000 (09:44 +0200)]
imx6: Add static PCA953x support in kernel
commit
a2386c384da501e862210f222994821580fe151c requires the
module to be static in the kernel.
So lets actually add support for it inside the kernel config.
Tested on Ventana
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Mathias Kresin [Thu, 8 Sep 2016 19:03:57 +0000 (21:03 +0200)]
ramips: set blocksize for remaining rt3883 devices
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Sat, 3 Sep 2016 13:27:46 +0000 (15:27 +0200)]
ramips: remove old build code seama recipe
Now that all seama images are using the new build code this seama recipe
used with the old build code can be dropped.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Thu, 8 Sep 2016 20:21:23 +0000 (22:21 +0200)]
ramips: switch DIR-610 A1 to new image build code
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Sat, 3 Sep 2016 08:18:31 +0000 (10:18 +0200)]
ramips: switch some rt3883 devices to new build code
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Sat, 3 Sep 2016 07:43:31 +0000 (09:43 +0200)]
ramips: move seama build recipe to Makefile
The identical build step is defined twice. Move it to the parent
Makefile to use only one version.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Fri, 2 Sep 2016 05:34:36 +0000 (07:34 +0200)]
ramips: switch DIR-615 H1 to new image build code
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Wed, 17 Aug 2016 05:49:34 +0000 (07:49 +0200)]
ramips: add build recipe for senao header
Add it to the Makefile to allow a board which will be supported soon
to use the recipe as well.
Signed-off-by: Mathias Kresin <dev@kresin.me>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
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>
Koen Vandeputte [Thu, 1 Sep 2016 14:10:01 +0000 (16:10 +0200)]
imx6: enable ARM crypto acceleration
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>
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>
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>
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>