Koen Vandeputte [Thu, 11 Jan 2018 11:55:02 +0000 (12:55 +0100)]
config: support new symbol intro'd in kernel 4.12
Symbol CONFIG_INITRAMFS_FORCE allows to ignore the value passed by the
bootloader.
By default, all symbols containing INITRAMFS are wiped from the final
config and then re-added conditionally.
Add support for this symbol, as the build will stop otherwise
questioning the user about this option:
* Restart config...
*
*
* General setup
*
Cross-compiler tool prefix (CROSS_COMPILE) []
Compile also drivers which will not load (COMPILE_TEST) [N/y/?] n
...
Initial RAM filesystem and RAM disk (initramfs/initrd) support
(BLK_DEV_INITRD) [Y/n/?] y
Initramfs source file(s) (INITRAMFS_SOURCE) []
Ignore the initramfs passed by the bootloader (INITRAMFS_FORCE)
[N/y/?] (NEW)
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Jo-Philipp Wich [Sat, 13 Jan 2018 13:46:11 +0000 (14:46 +0100)]
target: disable CONFIG_PROC_PAGE_MONITOR on most targets
Disable CONFIG_PROC_PAGE_MONITOR in most places and only keep it enabled
for virtual targets such as malta or potent ones like x86.
This saves up to 4KB of uncompressed kernel size and significantly
decreases CPU load under certain workloads.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Sven Eckelmann [Thu, 30 Nov 2017 13:30:06 +0000 (14:30 +0100)]
ipq806x: add support for OpenMesh A42
* QCA IPQ401x
* 256 MB of RAM
* 32 MB of SPI NOR flash (s25fl256s1)
- 2x 15 MB available; but one of the 15 MB regions is the recovery image
* 2T2R 2.4 GHz
- QCA4019 hw1.0 (SoC)
- requires special BDF in QCA4019/hw1.0/board-2.bin with
bus=ahb,bmi-chip-id=0,bmi-board-id=16,variant=OM-A42
* 2T2R 5 GHz
- QCA4019 hw1.0 (SoC)
- requires special BDF in QCA4019/hw1.0/board-2.bin with
bus=ahb,bmi-chip-id=0,bmi-board-id=17,variant=OM-A42
* multi-color LED (controlled via red/green/blue GPIOs)
* 1x button (reset; kmod-input-gpio-keys compatible)
* external watchdog
- triggered GPIO
* 1x USB (xHCI)
* TTL pins are on board (arrow points to VCC, then follows: GND, TX, RX)
* 2x gigabit ethernet
* powered only via POE
- 802.3af POE on Ethernet 1
- 18-24v passive POE (mode B) on Ethernet 2
The tool ap51-flash (https://github.com/ap51-flash/ap51-flash) should be
used to transfer the factory image to the u-boot when the device boots up.
Signed-off-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
Sven Eckelmann [Mon, 8 Jan 2018 17:28:51 +0000 (18:28 +0100)]
ipq806x: enable gpio watchdog support
The kernel driver gpio-wdt or the userspace tool om-watchdog can be used to
trigger external gpio watchdog chips. The gpio-wdt driver has the benefit
that it can be configured together with the rest of the device in the DTS
and better integrates in the OpenWrt via procd.
Signed-off-by: Sven Eckelmann <sven.eckelmann@openmesh.com>
Sven Eckelmann [Thu, 30 Nov 2017 13:30:06 +0000 (14:30 +0100)]
om-fwupgradecfg-gen.sh: add support for the A42
Signed-off-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
Sven Eckelmann [Thu, 30 Nov 2017 13:30:06 +0000 (14:30 +0100)]
build: add image command for CE images
Combined Extended Images V1 can be created easily via the new image
commands using
IMAGE/sysupgrade.bin/squashfs := append-rootfs | pad-rootfs | combined-ext-image
Signed-off-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
Sven Eckelmann [Thu, 30 Nov 2017 13:30:06 +0000 (14:30 +0100)]
combined-ext-image.sh: generate image in temp dir
The new build commands operate on the input image and use it again as
output image. This conflicts with the way combined-ext-image.sh was
operating. It required that input and output files are different files and
and that it can write freely to the output file.
This can be avoided when all intermediate build steps by
combined-ext-image.sh are done in a temporary directory. The output file is
then only overwritten in the last step.
Signed-off-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
Sven Eckelmann [Thu, 30 Nov 2017 13:30:06 +0000 (14:30 +0100)]
uboot-envtools: add OpenMesh A42 support
Signed-off-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
Sven Eckelmann [Thu, 30 Nov 2017 13:30:06 +0000 (14:30 +0100)]
ipq-wifi: add board-2.bin for OpenMesh A42
Signed-off-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
Sven Eckelmann [Thu, 30 Nov 2017 13:30:06 +0000 (14:30 +0100)]
mac80211: ath10k: search DT for BDF variant info
Board Data File (BDF) is loaded upon driver boot-up procedure. The right
board data file is identified on QCA4019 using bus, bmi-chip-id and
bmi-board-id.
The problem, however, can occur when the (default) board data file cannot
fulfill the vendor requirements and it is necessary to use a different
board data file.
This problem was solved for SMBIOS by adding a special SMBIOS type 0xF8.
Something similar has to be provided for systems without SMBIOS but with
device trees. No solution was specified by QCA and therefore a new one has
to be found for ath10k.
The device tree requires addition strings to define the variant name
wifi@
a000000 {
status = "okay";
qcom,ath10k-calibration-variant = "RT-AC58U";
};
wifi@
a800000 {
status = "okay";
qcom,ath10k-calibration-variant = "RT-AC58U";
};
This would create the boarddata identifiers for the board-2.bin search
* bus=ahb,bmi-chip-id=0,bmi-board-id=16,variant=RT-AC58U
* bus=ahb,bmi-chip-id=0,bmi-board-id=17,variant=RT-AC58U
Signed-off-by: Sven Eckelmann <sven.eckelmann@open-mesh.com>
Luis Araneda [Fri, 5 Jan 2018 13:11:16 +0000 (10:11 -0300)]
ipq806x: Sort occurrences of boardame alphabetically
This restores the alphabetical sort that was present
before the renaming of boardname.
Signed-off-by: Luis Araneda <luaraneda@gmail.com>
Luis Araneda [Fri, 5 Jan 2018 13:00:11 +0000 (10:00 -0300)]
ipq806x: convert to dt-based board-detection
Use the generic board detection method:
- Board name: First compatible string from the device tree
- Board model: Model property from the device tree
Change occurrences of board name in userspace by the compatible
string, and removed target specific board detection script
Replace the definition of SUPPORTED_DEVICES in Device/Default
to extract the dt compatible string from each device definition.
Additionally, for devices supported by lede-17.01, append
the value of BOARD_NAME to SUPPORTED_DEVICES in the device
definition.
Signed-off-by: Luis Araneda <luaraneda@gmail.com>
Luis Araneda [Sun, 31 Dec 2017 19:48:17 +0000 (16:48 -0300)]
ipq806x: sync image filename with dt compatible string
Use <manufacturer>_<modelname> as image name for board using the
devicetree compat string as boardname.
Signed-off-by: Luis Araneda <luaraneda@gmail.com>
Luis Araneda [Thu, 28 Dec 2017 02:38:50 +0000 (23:38 -0300)]
ipq806x: add a helper variable with the vendor name stripped
This will allow to maintain the current syntax for LEDs config
when switching to a device tree compatible string boardname.
None of the current boards use a comma in the boardname, so they
will be unaffected.
Signed-off-by: Luis Araneda <luaraneda@gmail.com>
Luis Araneda [Wed, 27 Dec 2017 02:39:39 +0000 (23:39 -0300)]
ipq806x: add missing dt compatible string to ap-dk04 board
Compatible string is mandatory to dts files
Signed-off-by: Luis Araneda <luaraneda@gmail.com>
Mathias Kresin [Mon, 8 Jan 2018 22:24:44 +0000 (23:24 +0100)]
ipq806x: drop partitial supported boards
There are only artifacts for these boards in our tree and not even
partial support.
Drop teh stale files.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Florian Fainelli [Fri, 12 Jan 2018 22:53:23 +0000 (14:53 -0800)]
armvirt: Resync kernel configuration
The 4.14 kernel configuration defaulted to a v4/v5 multiplatform while
4.9 was using a v6/v7 default configuration. Resync 4.14 against 4.9 so
they are nearly identical.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Florian Fainelli [Fri, 12 Jan 2018 22:49:55 +0000 (14:49 -0800)]
kernel: Add a bunch of missing symbols
Add a bunch of missing configuration symbols found while building
armvirt for 4.14 after re-synchronization of the configuration between
4.9 and 4.14.
Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
Rafał Miłecki [Fri, 12 Jan 2018 22:28:00 +0000 (23:28 +0100)]
bcm53xx: rename SPI driver fix to use 0xx prefix as it was accepted
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Rafał Miłecki [Fri, 12 Jan 2018 22:27:19 +0000 (23:27 +0100)]
bcm53xx: use upstream fix for PCI iproc driver regression
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Kevin Darbyshire-Bryant [Fri, 12 Jan 2018 11:23:41 +0000 (11:23 +0000)]
base-files: sysupgrade: correct command help text
Commit
30f61a34b4cfd2c676fea4a919e089d6a77254e9 claimed to drop -d & -p
options. In reality only -d was dropped. Update command help text to
reflect that -d is no longer a supported option.
Fixes FS#1187
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Rosen Penev [Wed, 10 Jan 2018 02:50:44 +0000 (18:50 -0800)]
ramips: fix cpuclock for the GB-PC1
The GnuBee PC1 stock bootloader runs at 900Mhz. This fixes bad clock drift when running the platform.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Christian Lamparter [Sun, 7 Jan 2018 20:27:52 +0000 (21:27 +0100)]
apm821xx: remove kernel 4.9 support
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Christian Lamparter [Sun, 7 Jan 2018 20:27:51 +0000 (21:27 +0100)]
apm821xx: switch to 4.14.x kernel
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Christian Lamparter [Sun, 7 Jan 2018 20:27:50 +0000 (21:27 +0100)]
apm821xx: backport crypto4xx patches from 4.15
This patch backports changes to crypto4xx in order
to get the crypto4xx operational.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Christian Lamparter [Sun, 7 Jan 2018 20:27:49 +0000 (21:27 +0100)]
apm821xx: convert MR24 to use DT PHY defintion
Convert the MR24 to use the DT phy probing and at803x PHY driver.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Christian Lamparter [Sun, 7 Jan 2018 20:27:48 +0000 (21:27 +0100)]
apm821xx: add linux 4.14 apm821xx patches
This patch updates the apm821xx target to use the 4.14 kernel.
4.14 finally ships with a driver for the WNDR4700's tc654 fan
controller. The custom driver is deprecated in favor of the
upstream driver and the thermal cooling definitions in the DTS
are updated.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Kevin Darbyshire-Bryant [Mon, 8 Jan 2018 12:59:46 +0000 (12:59 +0000)]
kmod-sched-cake: bump to latest cake bake
More important bug fix:
402f05c Use full-rate mtu_time in all tins. Fixes an issue where some
cake tins experienced excessive latency since
49776da (dynamically
adjust target)
Minor bug fixes:
31277c2 Avoid unsigned comparison against zero. Fix compiler warning,
no known impact.
8cf5278 ack_filter: fix TCP flag check. A very contrived case may have
lead to dropping a SYN packet that should not be dropped.
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Hauke Mehrtens [Mon, 8 Jan 2018 22:58:40 +0000 (23:58 +0100)]
at91: create device specific sdcard images
Without this patch one sdcard image with the following name is created
for all devices:
openwrt-at91-sama5--sdcard.img.gz
This makes the build system create device specific versions like:
openwrt-at91-sama5-at91-sama5d2_xplained-sdcard.img.gz
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Kevin Darbyshire-Bryant [Wed, 10 Jan 2018 14:06:52 +0000 (14:06 +0000)]
kernel: bump 4.14 to 4.14.13
Refresh patches
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Kevin Darbyshire-Bryant [Wed, 10 Jan 2018 14:07:17 +0000 (14:07 +0000)]
kernel: bump 4.4 to 4.4.111
Refresh patches
Tested-on: ar71xx Archer C7 v2
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Kevin Darbyshire-Bryant [Wed, 10 Jan 2018 14:07:41 +0000 (14:07 +0000)]
kernel: bump 4.9 to 4.9.76
Refresh patches
Tested-on: ar71xx Archer C7 v2
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Tested-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Jo-Philipp Wich [Wed, 10 Jan 2018 19:17:48 +0000 (20:17 +0100)]
build: fix restoring /etc/opkg with PER_DEVICE_ROOTFS
When generating per-device rootfs directories, the ./etc/opkg/ directory
is moved away prior to calling opkg install, opkg remove and rootfs_prepare.
After the opkg invocations and the rootfs_prepare macro call, the saved opkg
config directory is supposed to be moved back to its previous ./etc/opkg
location.
The mv command however can fail to properly restore the directory under
certain circumstances, e.g. when the prior opkg or files/ overlay copy
operations caused a new ./etc/opkg/ directory to be created.
In this case, the backed up directory (named target-dir-$hash.opkg) will be
moved into the preexisting ./etc/opkg/ directory instead, causing the opkg
configuration to be located in a wrong path on the final rootfs, e.g. in
/etc/opkg/target-dir-$hash.opkg/distfeeds.conf instead of
/etc/opkg/distfeeds.conf.
Solve this problem by replacing the naive "mv" command with a recursive
"cp -T" invocation which causes the backed up directory tree to get merged
with the destination directory in case it already exists.
Also perform the rootfs_prepare macro call after restoring the opkg
configuration, to allow users to override it again by using the files/
overlay mechanism.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Wed, 10 Jan 2018 22:53:45 +0000 (23:53 +0100)]
sdk: change base feed fallback to git.openwrt.org
Change the hardcoded lede-project.org base feed fallback to
$(PROJECT_GIT)/openwrt/openwrt.git instead.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Rafał Miłecki [Thu, 11 Jan 2018 12:59:32 +0000 (13:59 +0100)]
bcm53xx: replace linux,part-probe with a proper partitions subnode
This solution is more upstream compatible as it only requires specifying
of_match_table in the parser code and doesn't depend on linux,part-probe
which is solution made generic by a LEDE downstream patch that can't be
upstreamed.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Rafał Miłecki [Thu, 11 Jan 2018 10:59:39 +0000 (11:59 +0100)]
kernel: backport mtd implementation for "compatible" in "partitions" subnode
This backports upstream support for "compatible" DT property set for the
"partitions" subnode of flash node. It allows specifying how partitions
should be created/parsed. Right now only "fixed-partitions" is
supported.
It should eventually replace our downstream "linux,part-probe" solution.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Matthias Schiffer [Wed, 10 Jan 2018 14:38:42 +0000 (15:38 +0100)]
kernel: allow disabling multicast routing support
Multicast routing support is not needed in most setups, and increases the
size of the kernel considerably (>10K after LZMA). Add a config switch to
allow disabling it.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Matthias Schiffer [Tue, 9 Jan 2018 20:03:00 +0000 (21:03 +0100)]
generic: unlock Winbond flash on boot
The underlying issue breaking Spansion flash has been fixed with "mtd: spi-nor:
wait until lock/unlock operations are ready" and "mtd: spi-nor: wait for SR_WIP
to clear on initial unlock", so we can support unlocking for Winbond flash
again.
This is necessary to have writable flash on certain UBNT devices with some
bootloader versions.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Matthias Schiffer [Tue, 9 Jan 2018 19:05:04 +0000 (20:05 +0100)]
ebtables: update to latest git 2017-10-24
6a82659 Use flock() for --concurrent option
73c2371 ebtables: extensions: Constify option struct
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Jo-Philipp Wich [Thu, 11 Jan 2018 02:11:19 +0000 (03:11 +0100)]
sunxi: enable missing config symbol
Enable CONFIG_MFD_AXP20X_I2C after the switch to Kernel 4.9, this symbol
is reportedly required to get USB working on the Banana Pi.
Ref: https://forum.lede-project.org/t/sunxi-kernel-4-9-usb-not-working/10513
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Hans Dedecker [Mon, 8 Jan 2018 08:02:30 +0000 (09:02 +0100)]
omcproxy: silence fw3 warnings
Silence fw3 warnings in omcproxy init script in case fw3 is not enabled
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
Jo-Philipp Wich [Fri, 5 Jan 2018 09:46:06 +0000 (10:46 +0100)]
treewide: replace LEDE_GIT with PROJECT_GIT
Remove LEDE_GIT references in favor to the new name-agnostic
PROJECT_GIT variable.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Fri, 5 Jan 2018 09:43:55 +0000 (10:43 +0100)]
downloads.mk: introduce name-agnostic PROJECT_GIT variable
Introduce a name-agnostic PROJECT_GIT variable poiting to
https://git.openwrt.org/ and declare LEDE_GIT and OPENWRT_GIT
as aliases to it.
After some transition time we can drop this alias variables.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Mogula Pranay [Tue, 9 Jan 2018 06:13:29 +0000 (11:43 +0530)]
netifd: update dhcp.script to handle dynamic routing
Certain DHCP servers push a gateway outside of the assigned interface subnet,
to support those situations, install a host route towards the gateway.
If Gateway and IP are served in same network, openwrt quagga cannot learn
routes (rip routes are not getting added, showing inactive) whereas
working fine when Gateway and IP are in different network.
Signed-off-by: Mogula Pranay <mogula.pranay@nxp.com>
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
Mathias Kresin [Sat, 6 Jan 2018 17:38:35 +0000 (18:38 +0100)]
ltq-xdsl-app: drop script for renaming the netdev
Our netdevs are named dsl by default now, the rename via scripts isn't
required anymore.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Sat, 6 Jan 2018 16:30:38 +0000 (17:30 +0100)]
lantiq: create ATM/PTM interfaces with dsl as netdev name
Renaming an atm etherbride using 'ip link' (via hotplug) is racy since the
original netdev might disappear before br2684ctl has finished appling it's
setting:
local2.notice br2684ctl[1667]: Interface "nas0" created sucessfully
local2.notice br2684ctl[1667]: Communicating over ATM 0.8.35, encapsulation: LLC
kern.info kernel: dsl0: renamed from nas0
kern.err kernel: br2684:br2684_regvcc: tried to attach to non-existent device
local2.err br2684ctl[1667]: Could not configure interface:No such device or address
By passing the final used netdev name to br2684ctl_wrap another race
condition workaround will be enabled again.
Change the lantiq ptm driver to create a netdev with the name dsl as well.
Albeit the rename via 'ip link' works fine so far, using a different
approach for ptm then atm could be confusing.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Sat, 6 Jan 2018 16:37:13 +0000 (17:37 +0100)]
ltq-xdsl-app: drop manual br2684ctl reload
br2684ctl starts automatically, set up reload triggers, which fire as soon
as a atm driver is loaded. No need to do the reload via the script.
The reload is only required as soon as we can reliable switch between atm
and ptm driver and need to be implemented in a race free way.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Martin Schiller [Mon, 8 Jan 2018 12:12:20 +0000 (13:12 +0100)]
linux-atm: add br2684ctl option to specify the netdev name
Add the uci option nameprefix to specifc a target netdev name. Patch the
br2684ctl code to accept and set a netdev name via commandline parameters.
It allows to use the same netdev name for ATM and PTM lines on lantiq
xdsl hardware.
Signed-off-by: Martin Schiller <ms@dev.tdt.de>
Signed-off-by: Mathis Kresin <dev@kresin.me>
Jo-Philipp Wich [Tue, 9 Jan 2018 23:40:33 +0000 (00:40 +0100)]
armvirt: re-enable VFP support
Enable CONFIG_VFP again which was disabled during the 4.9 -> 4.14 bump.
Boot tested both 32 and 64 bit subtargets using initramfs images with the
qemu-system-arm and qemu-system-aarch64 emulators.
Fixes: aa100b66f2 ("armvirt: bump to v4.14")
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Tue, 9 Jan 2018 23:39:14 +0000 (00:39 +0100)]
armvirt: 64: add missing kernel config symbols
Fixes: aa100b66f2 ("armvirt: bump to v4.14")
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Kevin Darbyshire-Bryant [Sun, 7 Jan 2018 14:12:52 +0000 (14:12 +0000)]
kernel: bump 4.4 to 4.4.110
Refresh patches
Fixes: CVE-2017-5754 aka Meltdown
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
[fix typo in commit msg, conflict after 4.14 bump]
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Kevin Darbyshire-Bryant [Sun, 7 Jan 2018 14:08:03 +0000 (14:08 +0000)]
kernel: bump 4.9 to 4.9.75
Refresh patches
Fixes: CVE-2017-5754 aka Meltdown
Tested-on: ar71xx Archer C7 v2
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
Tested-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
[fix conflict after 4.14 bump]
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Hauke Mehrtens [Mon, 8 Jan 2018 22:42:31 +0000 (23:42 +0100)]
uboot-at91: make packages hidden
These packages are needed to generate the image, better mark them hidden
so we will activate them based on which boards gets build and they will
be activated always when the board which needs then gets build.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Mon, 8 Jan 2018 22:41:10 +0000 (23:41 +0100)]
at91bootstrap: make packages hidden
These packages are needed to generate the image, better mark them hidden
so we will activate them based on which boards gets build and they will
be activated always when the board which needs then gets build.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Hauke Mehrtens [Mon, 8 Jan 2018 22:20:57 +0000 (23:20 +0100)]
at91bootstrap: remove redundant definition of BUILD_SUBTARGET
It is sufficient to define this in AT91Bootstrap/Default as this is not
used for all.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Sandeep Sheriker Mallikarjun [Thu, 4 Jan 2018 19:47:17 +0000 (12:47 -0700)]
at91bootstrap: at91bootstrap default selection
The fixes following problems
1. changing prefix at91bootstrap to AT91bootstrap will fix the default
selection of at91bootstrap for the selected sama5 subtarget.
2. fixed missing default selection of sama5d4 nand flash for the
selected sama5d4 subtarget.
3. corrected at91bootstrap Title name.
Signed-off-by: Sandeep Sheriker Mallikarjun <sandeepsheriker.mallikarjun@microchip.com>
Sandeep Sheriker Mallikarjun [Thu, 4 Jan 2018 19:43:23 +0000 (12:43 -0700)]
uboot-at91: fix build after fpu activation
neon and VFPv4 support is added to this target and uboot-at91 build
fails due to TARGET_CFLAGS -mfloat-abi set to hard. as a fix, setting
uboot-at91 CFLAGS -mfloat-abi=soft.
Fixes: 01cc6bd495f ("at91: sama5: activate fpu")
Signed-off-by: Sandeep Sheriker Mallikarjun <sandeepsheriker.mallikarjun@microchip.com>
Jo-Philipp Wich [Tue, 9 Jan 2018 21:47:17 +0000 (22:47 +0100)]
kernel: kmod-w1 depends on kmod-hwmon-core since Linux 4.14
Fixes the following dependency error encountered by the buildbots:
Package kmod-w1 is missing dependencies for the following libraries:
hwmon.ko
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Mathias Kresin [Mon, 8 Jan 2018 22:49:22 +0000 (23:49 +0100)]
kirkwood: sort occurrences of boardame alphabetically
This restores the alphabetical sort that was present
before the renaming of boardname.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Alberto Bursi [Tue, 2 Jan 2018 15:32:01 +0000 (16:32 +0100)]
uboot-envtools: add pogoplug v4
add pogoplug v4 envs support in uboot-envtools
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
Alberto Bursi [Tue, 2 Jan 2018 15:07:19 +0000 (16:07 +0100)]
kirkwood: add pogoplug v4
also known as
POGO-V4-A3-02
or
POGO-V4-A3-01
SoC: Marvell
88F6192 800Mhz
SDRAM memory: 128MB
Gigabit ethernet: 1 Marvell
88E1310
Flash memory: 128MB
2 status LEDs (one green one red)
1 "Eject" button on the back (used as "Reset" button)
1 USB 2.0 port (on upper side)
1 sata slot (power + data) for 2.5'' drives (upper side)
2 USB 3.0 ports from a controller on PCIe x1 of the SoC
1 full-size SDcard slot (fits a whole SD card into it)
This device supports the (linux-only) kwboot tool to send
a new uboot over serial console, so it is easy to unbrick
in case the uboot is erased and the device won't boot.
-----
Install instructions:
-----
Since it's not possible to get ssh access to these
devices, the only way to take control is to
solder pins to get TTL serial access.
Case can be opened by removing screws beneath two rubber
feet at back of device, then lifting while prying the
sides of the upper part out to unhook a latch on each
side about 2/3rds of the way toward the front.
Serial connection pins are those labeled "J11", left
of SD as you face SD opening.
Pins are (from left to right, i.e. the first in the list
is the nearest to the SD slot) GND, Rx, Tx.
Do not connect +V pin if you use a USB (self-powered)
TTL-to-USB dongle. Any USB TTL-to-USB converter will work.
Baud rate is 115200, parity "none", databits "8",
flow control "none".
Stock uboot is unable to read ubi partitions (nor usb)
so we will replace it first with our uboot.
Start a TFTP server at IP address 169.254.254.254, and
place the uboot.kwb file in the folder of the server.
Start the serial session and then power up the device.
As soon as you see text on the serial start pressing random
letter keys to stop the boot process.
If you see something like the following line you can proceed:
CE>>
Otherwise if text is still scrolling by you missed the
opportunity, pull the plug and try again.
write
printenv ethaddr
The uboot will write something like this:
ethaddr=00:50:43:00:02:02
This is the device's MAC address, also present in the sticker
under the device.
Write this down as we will need to add it in the
new uboot configuration.
Use the following commands to load the new uboot:
tftp 0x20000 u-boot.kwb
If the uboot confirms that the transfer was successful,
then you can write it to flash with the following commands:
nand erase 0 0x200000
nand write 0x20000 0 0x1c0000
if after the last command the uboot wrote
"xxxx bytes written: OK"
then it was written correctly and we can proceed.
If it did not go well, try again or ask assistence in forums.
Shutting down or rebooting at this time will brick
the device, to unbrick it you will need to use the kwboot
tool from a Linux PC or Virtual Machine.
Now write:
reset
and press enter, the device will reboot and you should see
again text scrolling by.
Press a random key to stop it, and now you should see
pogoplugv4>
We now add the MAC address back, write:
setenv ethaddr '00:50:43:00:02:02'
Confirm that the uboot has understood by writing
printenv ethaddr
If all looks ok, save the setting with
saveenv
At this point the uboot is configured, and we only need to load
the firmware in the flash memory.
Follow the steps below in "Firmware recovery procedure".
----
Firmware recovery procedure
----
The new uboot allows easy recovery from a bad firmware upgrade
where you can't access the device anymore over ssh or luci.
Take a USB flash drive formatted as FAT32, and copy the
initramfs image file in it (it will have "initramfs" in the
file name), then rename it as "initramfs.bin".
Insert the USB drive in the USB 2.0 port of the pogoplug
(the port at the top).
Power up the device, and wait for it to finish booting.
The uboot should find and load the "initramfs.bin"
from usb and if you are connected with serial you should
see the linux kernel boot log (text scrolling by).
Once it is done, press Enter and you will be greeted by
the OpenWRT banner.
If you were not connected with serial just wait a bit and,
you will be able to access it with ssh or luci web interface
(once you find its IP).
The recovery "initramfs" images are run from RAM, so you will
have to do a normal sysupgrade (firmware upgrade) to write
a firmware image to flash memory.
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
Alberto Bursi [Tue, 2 Jan 2018 14:54:07 +0000 (15:54 +0100)]
uboot-kirkwood: add uboot for pogoplug v4
add an uboot able to boot a kernel in an ubi partition
This uboot also has a "recovery" feature, before
booting from flash it will try to boot a initramfs
image called "initramfs.bin" from a FAT32-formatted
USB drive connected to the USB 2.0 port.
(u-boot lacks drivers for usb 3.0 controllers)
Just rename the initramfs image and place it on
the usb drive, the uboot will load it.
In case there is no USB drive or no such file
is found, the uboot will boot from internal flash.
The whole check takes less than a second, boot times
are not impacted.
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
Mathias Kresin [Thu, 28 Dec 2017 09:39:19 +0000 (10:39 +0100)]
kirkwood: sync image filename with boardname
Use <manufacturer>_<modelname> as image name.
Use the BOARD_NAME variable to ensure that the former used boardname is
still used as the subdirectory name for the sysupgrade-tar image, to
not break sysupgrade from earlier versions.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Thu, 28 Dec 2017 09:18:48 +0000 (10:18 +0100)]
kirkwood: use the generic board detect
Drop the target specific detection function in favour of the generic
one provided by base-files.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Thu, 28 Dec 2017 08:22:45 +0000 (09:22 +0100)]
kirkwood: use image metadata
Append and enforce image metadata. Remove the device specific image
checks, they are replaced by image metadata.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Thu, 28 Dec 2017 00:15:57 +0000 (01:15 +0100)]
uboot-kirkwood: fix default u-boot selection
Build the Seagate GoFlexHome u-boot for the Seagate GoFlexNet as well. The
name clearly indicates that the u-boot can be used for both boards
Build the Zyxel NSA310 u-boot if the NSA310B image is selected.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Fri, 12 May 2017 19:01:44 +0000 (21:01 +0200)]
kirkwood: fix nsa310b compatible strings
Use a destinct compatible string fro the nsa310b.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Thu, 28 Dec 2017 09:07:48 +0000 (10:07 +0100)]
kirkwood: fix compatible string in ib62x0 dts
There is no point in being that specifc in a generic dts covering a range
of boards. Keep only the the generic compat string to use it for a
devicetree based boardname.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Wed, 27 Dec 2017 23:40:49 +0000 (00:40 +0100)]
kirkwood: drop support for orphaned boards
No image build code for the Guruplug, Sheevaplug and NSA310S exists. Drop
support for the boards for now.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Wed, 27 Dec 2017 23:25:41 +0000 (00:25 +0100)]
kirkwood: cleanup image build code
Drop NAND_BLOCKSIZE, UBI_OPTS and UBIFS_OPTS. They are either used by
not supported filesystems or by the legacy image build code.
Add common used options/images to the default build code and override
the options where necessary.
Don't export the kernel image, it isn't required by any board.
While at it, change the file extension for the sysupgrade images to bin.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Mon, 8 Jan 2018 22:27:35 +0000 (23:27 +0100)]
ipq806x: drop kernel 4.4 leftovers
Remove the kernel 4.4 specific devicetree source files.
Fixes: 3a3564ead5e4 ("ipq806x: remove v4.4 support")
Signed-off-by: Mathias Kresin <dev@kresin.me>
David Bauer [Tue, 19 Dec 2017 01:32:47 +0000 (02:32 +0100)]
ar71xx: fix Archer C7v4 5GHz MAC-address
The TP-Link firmware uses (primary_mac-1) as MAC-address
for the 5GHz WiFi. This applies the same behaviour to LEDE.
Currently, the MAC-address is retrieved from eth1, which
does not exist on the Archer C7 v4. As a result from this,
every C7 v4 with LEDE carries the same MAC-Address on the 5GHz WiFi.
Signed-off-by: David Bauer <mail@david-bauer.net>
Chuanhong Guo [Fri, 29 Dec 2017 04:35:32 +0000 (12:35 +0800)]
ramips: fix lenovo newifi-y1 switch and LED config
There are 3 ethernet ports on Y1. LAN1 on port1, LAN2 on port0 and WAN on
port4.
Use a standalone switch configuration to match this and use the switch
trigger so that LAN LED could indicate the connetction status for both
lan ports correctly.
This patch also drop the internet led configuration, because there is a
WAN led for port4 and eth0.2 isn't always used as WAN.
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
Mathias Kresin [Sat, 30 Dec 2017 19:41:22 +0000 (20:41 +0100)]
ramips: add missing soc compatibles to dts
Add SoC compatibles where missing.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Sat, 30 Dec 2017 09:19:18 +0000 (10:19 +0100)]
ramips: firewrt: indicate boot status via LED
Add the Firefly FireWRT gree power LED to diag.sh to indicate the boot
status via the power LED.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Thu, 21 Dec 2017 00:06:42 +0000 (01:06 +0100)]
base-files: gpio switch: set output value with direction
Use the "low" and "high" values to configure the GPIO as an output with
that initial value. It ensures that the gpio doesn't have a unwanted value
during the time the direction is set to ouput and the actual value is
applied.
We don't need to take care of the GPIO polarity for now, since our
exported GPIOs are always active low.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Mathias Kresin [Thu, 21 Dec 2017 08:44:53 +0000 (09:44 +0100)]
base-files: gpio_switch: start before boot state done is set
Start gpio_switch before the boot state is set to up/initialised/done.
This way the exported GPIOs are available at the time rc.local is started.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Gabor Juhos [Tue, 9 Jan 2018 20:20:14 +0000 (21:20 +0100)]
ar71xx: remove obsolete '< KERNEL_VERSION(4, 4, 0)' check
Because the oldest supported kernel version on the ar71xx target is 4.4,
the condition that the kernel version is smaller than 4.4.0 is always
false. Remove the obsolete check from mach-rb4xx.c to clean up the code
a bit.
Signed-off-by: Gabor Juhos <juhosg@freemail.hu>
Gabor Juhos [Tue, 9 Jan 2018 20:20:13 +0000 (21:20 +0100)]
ar71xx: remove obsolete '< KERNEL_VERSION(4, 2, 0)' check
Because the oldest supported kernel version on the ar71xx target is 4.4,
the condition that the kernel version is smaller than 4.2.0 is always
false. Remove the obsolete check from ag71xx_main.c to clean up the code
a bit.
Signed-off-by: Gabor Juhos <juhosg@freemail.hu>
Gabor Juhos [Tue, 9 Jan 2018 20:20:12 +0000 (21:20 +0100)]
ar71xx: remove obsolete '< KERNEL_VERSION(3, 15, 0)' checks
Because the oldest supported kernel version on the ar71xx target is 4.4,
the condition that the kernel version is smaller than 3.15.0 is always
false. Remove the obsolete checks from the target specific source files
to clean up the code a bit.
Signed-off-by: Gabor Juhos <juhosg@freemail.hu>
John Crispin [Tue, 9 Jan 2018 13:40:54 +0000 (14:40 +0100)]
armvirt: bump to v4.14
Signed-off-by: John Crispin <john@phrozen.org>
John Crispin [Tue, 9 Jan 2018 09:25:37 +0000 (10:25 +0100)]
malta: bump tp v4.14
Signed-off-by: John Crispin <john@phrozen.org>
John Crispin [Tue, 9 Jan 2018 09:22:54 +0000 (10:22 +0100)]
malta: fix image prefix names
Signed-off-by: John Crispin <john@phrozen.org>
Gabor Juhos [Tue, 9 Jan 2018 07:18:16 +0000 (08:18 +0100)]
ar71xx/mikrotik: disable TP-Link TL-WR810N v2 board support
It does not belongs to the MikroTik boards.
Signed-off-by: Gabor Juhos <juhosg@freemail.hu>
Gabor Juhos [Tue, 9 Jan 2018 07:18:15 +0000 (08:18 +0100)]
ar71xx/mikrotik: disable unused MTD partitioning and split drivers
Reduces the compressed kernel size by ~2.5KB.
Signed-off-by: Gabor Juhos <juhosg@freemail.hu>
Gabor Juhos [Tue, 9 Jan 2018 07:18:14 +0000 (08:18 +0100)]
ar71xx/mikrotik: disable unused ethernet switch and phy drivers
Disable the drivers for the following ethernet switches:
Realtek RTL8306
Realtek RTL8366/8367
Marvell
88E6060 (DSA)
Marvell
88E6063 (DSA)
Also disable the phy driver for Marvell PHYs.
None of the supported RouterBOARDs are using any device
which needs these drivers.
Disable the DSA framework too, because it is not required
by the remaining switch drivers.
This reduces the compressed kernel size by ~20KB.
Signed-off-by: Gabor Juhos <juhosg@freemail.hu>
Gabor Juhos [Mon, 8 Jan 2018 14:47:49 +0000 (15:47 +0100)]
ar71xx: fix Kconfig dependency of the
88E6063 switch DSA driver
The Marvell
88E6063 ethernet switch driver depends on the DSA
framework. Add a 'depends on NET_DSA' statement to its Kconfig
entry to state that explicitly.
Fixes the following Kconfig warning:
warning: (NET_DSA_MV88E6060 && NET_DSA_MV88E6063) selects NET_DSA_TAG_TRAILER which has unmet direct dependencies (NET && NET_DSA)
Signed-off-by: Gabor Juhos <juhosg@freemail.hu>
Gabor Juhos [Mon, 8 Jan 2018 14:47:11 +0000 (15:47 +0100)]
ar71xx: sort kernel configurations
The symbols in some kernel configurations of the target are in
wrong order. Sort them with kconfig.pl.
Signed-off-by: Gabor Juhos <juhosg@freemail.hu>
Stijn Tintel [Mon, 8 Jan 2018 18:09:38 +0000 (20:09 +0200)]
octeon: add support for kernel 4.14
Tested on EdgeRouter Lite.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Stijn Tintel [Mon, 8 Jan 2018 11:40:26 +0000 (13:40 +0200)]
kernel: bump 4.14 to 4.14.12
No patch refresh required.
Compile-tested on: octeon, x86/64.
Runtime-tested on: octeon, x86/64.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Hauke Mehrtens [Sat, 6 Jan 2018 01:36:39 +0000 (02:36 +0100)]
lantiq: activate noise margin delta for VDSL too
Previously this was only activated for ADSL, this patch activates the
same setting also for VDSL, this feature is also support for VDSL in the
same way it works for ADSL.
I tested it with DSL FW 5.7.9.5.1.7 against a Broadcom 177.140 DSLCO
(Deutsche Telekom) and saw different data rates and Max. Attainable Data
Rates depending on the ds_snr_offset settings I choose.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Matthias Schiffer [Sun, 7 Jan 2018 18:52:05 +0000 (19:52 +0100)]
uboot-envtool: remove superfluous + sign from PKG_BUILD_DEPENDS
A + sign does not have meaning in build depends.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Jo-Philipp Wich [Sun, 7 Jan 2018 15:33:51 +0000 (16:33 +0100)]
libubox: fix package bump
The previous commit was incorrectly rebased and referred to a not
yet existing PROJECT_GIT variable.
Fixes: d86a269c1f libubox: update to latest git HEAD
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Sun, 7 Jan 2018 15:15:52 +0000 (16:15 +0100)]
libubox: update to latest git HEAD
1c08e80 jshn: properly support JSON "null" type
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Sun, 7 Jan 2018 12:42:36 +0000 (13:42 +0100)]
Revert "fstools: update to latest git HEAD"
This reverts commit
df326e9b2e71a2c639371ef8acec0ac98bc36122.
The referenced commit does not exist in the upstream repository.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
John Crispin [Sun, 7 Jan 2018 12:02:38 +0000 (13:02 +0100)]
fstools: update to latest git HEAD
2a9a6ea libfstools: optimize building directory string for glob
de6b026 libfstools: support file paths longer than 255 chars
Signed-off-by: John Crispin <john@phrozen.org>
Gabor Juhos [Fri, 5 Jan 2018 18:13:20 +0000 (19:13 +0100)]
ar71xx: fix compiler warnings in mdio-bitbang.c for 4.4
Building Linux 4.4 for ar71xx throws the following warnings in
drivers/net/phy/mdio-bitbang.c:
CC drivers/net/phy/mdio-bitbang.o
In file included from include/linux/irqflags.h:14:0,
from drivers/net/phy/mdio-bitbang.c:20:
drivers/net/phy/mdio-bitbang.c: In function 'mdiobb_read':
include/linux/typecheck.h:11:18: warning: comparison of distinct pointer types lacks a cast
(void)(&__dummy == &__dummy2); \
^
include/linux/irqflags.h:63:3: note: in expansion of macro 'typecheck'
typecheck(unsigned long, flags); \
^
include/linux/irqflags.h:124:3: note: in expansion of macro 'raw_local_irq_save'
raw_local_irq_save(flags); \
^
drivers/net/phy/mdio-bitbang.c:162:2: note: in expansion of macro 'local_irq_save'
local_irq_save(flags);
^
include/linux/typecheck.h:11:18: warning: comparison of distinct pointer types lacks a cast
(void)(&__dummy == &__dummy2); \
^
include/linux/irqflags.h:68:3: note: in expansion of macro 'typecheck'
typecheck(unsigned long, flags); \
^
include/linux/irqflags.h:126:39: note: in expansion of macro 'raw_local_irq_restore'
#define local_irq_restore(flags) do { raw_local_irq_restore(flags); } while (0)
^
drivers/net/phy/mdio-bitbang.c:175:2: note: in expansion of macro 'local_irq_restore'
local_irq_restore(flags);
^
drivers/net/phy/mdio-bitbang.c:159:11: warning: unused variable 'i' [-Wunused-variable]
int ret, i;
^
In file included from include/linux/irqflags.h:14:0,
from drivers/net/phy/mdio-bitbang.c:20:
drivers/net/phy/mdio-bitbang.c: In function 'mdiobb_write':
include/linux/typecheck.h:11:18: warning: comparison of distinct pointer types lacks a cast
(void)(&__dummy == &__dummy2); \
^
include/linux/irqflags.h:63:3: note: in expansion of macro 'typecheck'
typecheck(unsigned long, flags); \
^
include/linux/irqflags.h:124:3: note: in expansion of macro 'raw_local_irq_save'
raw_local_irq_save(flags); \
^
drivers/net/phy/mdio-bitbang.c:185:2: note: in expansion of macro 'local_irq_save'
local_irq_save(flags);
^
include/linux/typecheck.h:11:18: warning: comparison of distinct pointer types lacks a cast
(void)(&__dummy == &__dummy2); \
^
include/linux/irqflags.h:68:3: note: in expansion of macro 'typecheck'
typecheck(unsigned long, flags); \
^
include/linux/irqflags.h:126:39: note: in expansion of macro 'raw_local_irq_restore'
#define local_irq_restore(flags) do { raw_local_irq_restore(flags); } while (0)
^
drivers/net/phy/mdio-bitbang.c:200:2: note: in expansion of macro 'local_irq_restore'
local_irq_restore(flags);
^
These are caused by the 900-mdio_bitbang_ignore_ta_value and
901-phy-mdio-bitbang-prevent-rescheduling-during-command patches.
The first patch removes some code but it does not remove the variable
which is used by the removed code only.
The second patch adds local_irq_{save,restore} calls. The type of the
argument of these calls must be 'unsigned long', but the patch defines
the variable as 'long'.
Fix both patches to silence the warnings.
Signed-off-by: Gabor Juhos <juhosg@freemail.hu>
Jo-Philipp Wich [Sun, 7 Jan 2018 11:42:14 +0000 (12:42 +0100)]
hostapd: bump PKG_RELEASE after 802.11w changes
Fixes: 8a57531855 "hostapd: set group_mgmt_cipher when ieee80211w is enabled"
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Wed, 3 Jan 2018 12:13:10 +0000 (13:13 +0100)]
kernel: fix packaging of kmod-gpio-nxp-74hc164
The NXP 74HC164 GPIO expander driver uses a different config symbol
("CONFIG_GPIO_74X164") and module name since since at least Kernel
version 2.6.37.
Update the kmod package definition accordingly by adjusting kconfig
and module file names.
This unrelated, but correct change has been separated from the
WNR2000v5 support commits.
Ref: https://github.com/lede-project/source/pull/1256
Suggested-by: Raphael Catolino <raphael.catolino@gmail.com>
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Jo-Philipp Wich [Thu, 4 Jan 2018 13:43:34 +0000 (14:43 +0100)]
zlib: only enable NEON optimizations on eligible targets
Instead of inferring the availability of NEON support from the target
optimization flags, use a preprocessor test to decide whether to enable
ARMv8 NEON optimizations.
Fixes the following build error spotted by the mediatek/32 buildbot:
[ 26%] Building C object CMakeFiles/zlib.dir/contrib/arm/inflate.o
In file included from .../zlib-1.2.11/contrib/arm/chunkcopy.h:10:0,
from .../zlib-1.2.11/contrib/arm/inflate.c:87:
.../arm_neon.h:31:2: error: #error You must enable NEON instructions (e.g. -mfloat-abi=softfp -mfpu=neon) to use arm_neon.h
#error You must enable NEON instructions (e.g. -mfloat-abi=softfp -mfpu=neon) to use arm_neon.h
^
In file included from .../zlib-1.2.11/contrib/arm/inflate.c:87:0:
.../zlib-1.2.11/contrib/arm/chunkcopy.h:18:9: error: unknown type name 'uint8x16_t'
typedef uint8x16_t chunkcopy_chunk_t;
^
[...]
CMakeFiles/zlib.dir/build.make:302: recipe for target 'CMakeFiles/zlib.dir/contrib/arm/inflate.o' failed
Fixes: 3acecba520 "package/libs/zlib: Add ARM and NEON optimizations"
Signed-off-by: Jo-Philipp Wich <jo@mein.io>