openwrt/staging/xback.git
6 years agobcm53xx: add DT patch describing pins mux controller
Rafał Miłecki [Fri, 9 Nov 2018 21:28:31 +0000 (22:28 +0100)]
bcm53xx: add DT patch describing pins mux controller

It's needed to support new devices that use specific pin functions.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomt76: update to the latest version
Felix Fietkau [Fri, 9 Nov 2018 14:26:05 +0000 (15:26 +0100)]
mt76: update to the latest version

ffccb48 mt76: do not wake tx queues during flush

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agokernel: deprecate support for the generic "linux,part-probe" DT binding
Rafał Miłecki [Fri, 9 Nov 2018 13:57:45 +0000 (14:57 +0100)]
kernel: deprecate support for the generic "linux,part-probe" DT binding

It has been rejected upstream and instead a nice/more generic solution
has been implemented. It's possible now to describe partitions format
using "compatible" DT string.

No OpenWrt target uses "linux,part-probe" anymore, leave it only in case
some forks need it. It will be dropped with support for new kernels.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agotools: firmware-utils: mksercommfw build on Darwin
Kevin Darbyshire-Bryant [Thu, 8 Nov 2018 21:35:17 +0000 (21:35 +0000)]
tools: firmware-utils: mksercommfw build on Darwin

asm/byteorder.h & hence __cpu_to_be32() doesn't exist on Darwin

Shamelessly copy some byte swap functions from oseama.c

Acked-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
6 years agoramips: use separated mtd patch reverting write_page() removal
Rafał Miłecki [Fri, 9 Nov 2018 12:30:51 +0000 (13:30 +0100)]
ramips: use separated mtd patch reverting write_page() removal

This helps understanding mtd changes needed to get downstream NAND
driver working.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agopopt: Add main site back to PKG_SOURCE_URL
Daniel Engberg [Wed, 7 Nov 2018 21:44:36 +0000 (22:44 +0100)]
popt: Add main site back to PKG_SOURCE_URL

Main site is back up after weeks of downtime.
Add it back as last resort

Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
6 years agobcm53xx: update pinctrl driver
Rafał Miłecki [Fri, 9 Nov 2018 06:40:42 +0000 (07:40 +0100)]
bcm53xx: update pinctrl driver

It's upstream now with a one trivial fix.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agoprocd: update to latest git HEAD
Hans Dedecker [Thu, 8 Nov 2018 13:46:56 +0000 (14:46 +0100)]
procd: update to latest git HEAD

20192e4 watchdog: improve seting watchdog timeout and frequency

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agooxnas: remove superseded sysupgrade image check
Daniel Golle [Mon, 5 Nov 2018 18:04:48 +0000 (19:04 +0100)]
oxnas: remove superseded sysupgrade image check

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
6 years agooxnas: kd20: correct memory size to 256MB
Daniel Golle [Mon, 5 Nov 2018 17:17:35 +0000 (18:17 +0100)]
oxnas: kd20: correct memory size to 256MB

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
6 years agoramips: fix netgear r6120 factory image generation
Koen Vandeputte [Thu, 8 Nov 2018 11:46:37 +0000 (12:46 +0100)]
ramips: fix netgear r6120 factory image generation

as indicated in commit c5bf408ed6bd "(ramips: fix image generation for mt76x8")
more rework was needed to fix the other issues.

Building on another machine, but using the same arch, showed
the application failing again for different reasons.

Fix this by completely rewriting the application, fixing following found issues:

- buffer overflows, resulting in stack corruption
- flaws in memory requirement calculations (too small, too large)
- memory leaks
- missing bounds checking on string handling
- non-reproducable images, by using unitilized memory in checksum calculation
- missing error handling, resulting in succes on specific image errors
- endianness errors when building on BE machines
- various minor build warnings
- documentation did not match the code actions (header item locations)
- allowing input to be decimal, hex or octal now

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoscript: ipkg-build: honour $SOURCE_DATE_EPOCH
Jo-Philipp Wich [Thu, 8 Nov 2018 10:52:33 +0000 (11:52 +0100)]
script: ipkg-build: honour $SOURCE_DATE_EPOCH

When the SOURCE_DATE_EPOCH environment variable is set, use it to
override the timestamps of .ipk archive contents.

This ensures that .ipk archives built in environments without SCM
metadata (mainly the SDK) are reproducible between different runs.

Ref: https://github.com/openwrt/packages/issues/6954
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
6 years agoiproute2: update to 4.19.0
Hans Dedecker [Mon, 5 Nov 2018 14:32:48 +0000 (15:32 +0100)]
iproute2: update to 4.19.0

Update to the latest version of iproute2; see https://lwn.net/Articles/769354/
for a full overview of the changes in 4.19.
Remove 190-add-cake-to-tc patch as CAKE qdisc is now supported in 4.19.0

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agomt76: update to the latest version, fix a typo
Felix Fietkau [Thu, 8 Nov 2018 08:17:18 +0000 (09:17 +0100)]
mt76: update to the latest version, fix a typo

29d9532 mt7603: fix faulty variable initialization in last commit

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agomt76: update to the latest version
Felix Fietkau [Wed, 7 Nov 2018 19:35:23 +0000 (20:35 +0100)]
mt76: update to the latest version

417cf49 mt7603: improve recovery from MCU hang

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agomac80211: brcmfmac: backport the last accepted 4.21 changes
Rafał Miłecki [Wed, 7 Nov 2018 10:16:25 +0000 (11:16 +0100)]
mac80211: brcmfmac: backport the last accepted 4.21 changes

It's a typo fix & patch that helps debugging possible WARN-ings.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: brcmfmac: backport NVRAM loading improvements
Rafał Miłecki [Wed, 7 Nov 2018 10:06:16 +0000 (11:06 +0100)]
mac80211: brcmfmac: backport NVRAM loading improvements

This adds support for storing board specific NVRAM files as firmware.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: brcmfmac: backport firmware loading cleanup
Rafał Miłecki [Wed, 7 Nov 2018 08:43:14 +0000 (09:43 +0100)]
mac80211: brcmfmac: backport firmware loading cleanup

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: brcmfmac: backport the latest 4.20 changes
Rafał Miłecki [Wed, 7 Nov 2018 08:01:32 +0000 (09:01 +0100)]
mac80211: brcmfmac: backport the latest 4.20 changes

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomac80211: brcmfmac: rename 4.20 backport patches
Rafał Miłecki [Wed, 7 Nov 2018 07:52:46 +0000 (08:52 +0100)]
mac80211: brcmfmac: rename 4.20 backport patches

Include kernel version to help tracking changes.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agomt76: update to the latest version
Felix Fietkau [Tue, 6 Nov 2018 21:25:06 +0000 (22:25 +0100)]
mt76: update to the latest version

172f081 mt76x0: do not overwrite other MT_BBP(AGC, 8) fields
2ae2046 mt76x0: phy: use mt76_poll_msec in mt76x0_phy_temp_sensor
add66a0 mt76x0: init: use mt76x02_mac_shared_key_setup in mt76x0_init_hardware
1f24db9 mt76x2: move wcid_tx_rate conf at bootstrap
097b2bd mt76x0: init: use mt76x02_mac_wcid_setup for wcid configuration
d0eb03e mt76x2u: init: remove not useful configuration
c7aa8f4 mt76x2u: init: use common routines for wcid/key initialization
bd568d9 mt76: move mt76x02_eeprom_copy in mt76x02-lib module
57b441d mt76x0: phy: introduce tssi calibration support
a78802c mt76x0: phy: use tssi reported value to configure tx power if available
94ad030 mt76: remove mcu_msg_alloc
15cb426 mt76: remove wait argument from mt76x02_mcu_function_select
cd76a83 mt76: remove wait argument from mt76x02_mcu_set_radio_state
9b97e0f mt76x02: run calibration after scanning
663b197 mt76x02: assure we update gain after scan
eb53745 mt76x0: dfs: fix IBI_R11 configuration on non-radar channels
ef06c30 mt76: avoid queue/status spinlocks while passing tx status to mac80211
5195257 mt7603: fix maximum rx frame size
01815e9 mt7603: use mutex from struct mt76_dev
3c8e8a7 mt7603: use irq_lock/irqmask from struct mt76_dev
8b7f50d mt76: add size check for additional rx fragments
87bad3a mt7603: stop MAC on watchdog reset
c34baab mt7603: add missing unlock in case changing the channel fails
ac6fc78 mt7603: add debugfs file to trigger watchdog resets
d9b5324 mt7603: add SPDX license identifiers
498d716 mt7603: fix checkpatch issues
a1afefd mt7603: fix handling of fixed rate packets
5ee0523 mt7603: fix tx status rate reporting for fixed-rate packets

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agoramips: add missing DTS frequency limit for the 5 GHz radio on Archer C20i
Felix Fietkau [Tue, 6 Nov 2018 21:24:03 +0000 (22:24 +0100)]
ramips: add missing DTS frequency limit for the 5 GHz radio on Archer C20i

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agovalgrind: bump to 4.14.0
Koen Vandeputte [Tue, 6 Nov 2018 15:33:09 +0000 (16:33 +0100)]
valgrind: bump to 4.14.0

Release notes can be found here:

http://www.valgrind.org/docs/manual/dist.news.html

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoramips: fix image generation for mt76x8
Koen Vandeputte [Tue, 6 Nov 2018 14:10:17 +0000 (15:10 +0100)]
ramips: fix image generation for mt76x8

Buildbot fails to generate images for targets also generating a
Sercomm binary with following error:

Opening file: /mnt/ramdisk/koen/firmware/builds/owrt_mt76x8/build_dir/target-mipsel_24kc_musl/linux-ramips_mt76x8/tmp/openwrt-ramips-mt76x8-netgear_r6120-squashfs-factory.img.rootfs.zip
Filesize: 3648606 .
mksercommfw: malloc.c:2427: sysmalloc: Assertion `(old_top == initial_top (av) && old_size == 0) || ((unsigned long) (old_size) >= MINSIZE && prev_inuse (old_top) && ((unsigned long) old_end & (pagesize - 1)) == 0)' failed.
Makefile:287: recipe for target '/mnt/ramdisk/koen/firmware/builds/owrt_mt76x8/build_dir/target-mipsel_24kc_musl/linux-ramips_mt76x8/tmp/openwrt-ramips-mt76x8-netgear_r6120-squashfs-factory.img' failed

Debugging using valgrind shows stack corruption due to a buffer overflow.
The author of the generator assumes the filename ends with "root",
while it should be "rootfs".

Fix this by accounting for the 2 missing characters which solves the build issues.

More work is required to cleanup this source, which will be done later on.

Reported-by: Hannu Nyman <hannu.nyman@iki.fi>
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoimx6: fix DMA transaction errors
Koen Vandeputte [Mon, 5 Nov 2018 16:41:00 +0000 (17:41 +0100)]
imx6: fix DMA transaction errors

Following errors were seen in the past on imx6 when using serial:

[ 22.617622] imx-uart 2020000.serial: DMA transaction error.
[ 22.623228] imx-uart 2020000.serial: DMA transaction error.
[ 22.628826] imx-uart 2020000.serial: DMA transaction error.
[ 22.648951] imx-uart 2020000.serial: DMA transaction error.
[ 22.654558] imx-uart 2020000.serial: DMA transaction error.
[ 22.660156] imx-uart 2020000.serial: DMA transaction error.

Which is the reason why DMA for the serial ports
got disabled in commits:

efb362cd93b0 ("imx6: disable dma on uart")
3b4241071dd4 ("imx6: disable UART dma")

As indicated on mailinglist discussion, the cause seems to be
the usage of very old SDMA firmware which is present in the soc:

[    0.624302] imx-sdma 20ec000.sdma: Direct firmware load for imx/sdma/sdma-imx6q.bin failed with error -2
[    0.624318] imx-sdma 20ec000.sdma: Falling back to user helper
[   64.531607] imx-sdma 20ec000.sdma: external firmware not found, using ROM firmware

This patch adds the new firmware binary. (2196 bytes)

It is required to embed the binary into the kernel image, as it
gets loaded very early in the boot process where the rootfs is not
available yet:

[    0.622966] imx-sdma 20ec000.sdma: loaded firmware 3.3

Extended testing shows that the DMA errors are not seen anymore
when using this newer firmware version.

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoath9k: fix dynack in IBSS mode
Koen Vandeputte [Mon, 5 Nov 2018 10:31:12 +0000 (11:31 +0100)]
ath9k: fix dynack in IBSS mode

Currently, dynack was only tested upstream using AP/STA mode.
Testing it on IBSS, showed that late-ack detection was broken.

This is caused due to dynack using Association Request/Response
frames for late-ack detection, which IBSS does not use.
Also allowing Authentication frames here solves this.

A second issue also got fixed, which was also seen AP/STA mode:

When a station was added, the estimated value would be exponentially averaged
using 0 as a starting point.

This means that on larger distances, the ack timeout was still not high
enough before synchronizing would run out of late-ack's for estimation.

Fix this by using the initial estimated value as a baseline
and only start averaging in the following estimation rounds.

Test setup:
- 2x identical devices:  RB912UAG-5HPnD + 19dB sector
- IBSS
- 2x2 802.11an (ar9340), HT20, long GI
- RSSI's  -70 / -71
- Real distance: 23910 meter

Results (60s iperf runs):

Fixed coverage class 54 (up to 24300m):
* 21.5 Mbits/sec

Dynack:
* 28.9 Mbits/sec

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoar71xx: disable some tiny targets with kernel 4.14
Koen Vandeputte [Mon, 5 Nov 2018 11:53:55 +0000 (12:53 +0100)]
ar71xx: disable some tiny targets with kernel 4.14

The bump to kernel 4.14 caused a massive increase in kernel size.

For most targets, switching them to dynamic partitioning allowed
to cope with this.

On some targets, the kernel partition is located behind the rootfs,
which disallows switching to dynamic partitioning as the boot location
would be altered, requiring a u-boot change.

Also within the tiny section, which disables kernel symbols etc
to decrease the image size,  the partition size is still too small.

Disable these targets for now, fixing image generation:

- Buffalo BHR-4GRV2
- Zbtlink ZBT-WE1526

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agokernel: bump 4.14 to 4.14.79
Koen Vandeputte [Mon, 5 Nov 2018 10:22:43 +0000 (11:22 +0100)]
kernel: bump 4.14 to 4.14.79

Refreshed all patches.

Compile-tested on: ar71xx, cns3xxx, imx6, x86_64
Runtime-tested on: ar71xx, cns3xxx, imx6, x86_64

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoglibc: update to latest 2.27 commit (BZ#23709,BZ#23822)
Hans Dedecker [Mon, 5 Nov 2018 13:19:33 +0000 (14:19 +0100)]
glibc: update to latest 2.27 commit (BZ#23709,BZ#23822)

d8eee5ef55 x86: Fix Haswell CPU string flags (BZ#23709)
5cd5309d91 i64: fix missing exp2f, log2f and powf symbols in libm.a [BZ #23822]
1759ea197b conform: XFAIL siginfo_t si_band test on sparc64

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agotools/bison: update to version 3.2
Hannu Nyman [Sun, 4 Nov 2018 15:42:09 +0000 (17:42 +0200)]
tools/bison: update to version 3.2

Update bison to version 3.2

Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
6 years agoipq40xx: fix NBG6617 LED mapping
David Bauer [Sun, 4 Nov 2018 15:01:39 +0000 (16:01 +0100)]
ipq40xx: fix NBG6617 LED mapping

The NBG6617's LEDs are wrongly identified in the 01_leds boardinit
script (board instead of boardname), resulting in referencing
non-existent LEDs in UCI.

Signed-off-by: David Bauer <mail@david-bauer.net>
6 years agooxnas: add console=ttyS0,115200 argument to bootargs
Daniel Golle [Sun, 4 Nov 2018 17:26:28 +0000 (18:26 +0100)]
oxnas: add console=ttyS0,115200 argument to bootargs

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
6 years agooxnas: enable image metadata by setting SUPPORTED_DEVICES
Daniel Golle [Sun, 4 Nov 2018 17:24:57 +0000 (18:24 +0100)]
oxnas: enable image metadata by setting SUPPORTED_DEVICES

For each board, set the legacy name as well as the new based on the
compatible-string from devicetree.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
6 years agooxnas: remove stray kernel config symbols
Daniel Golle [Sun, 4 Nov 2018 17:23:51 +0000 (18:23 +0100)]
oxnas: remove stray kernel config symbols

CONFIG_CRYPTO_* shouldn't be set by the target

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
6 years agooxnas: switch to generic board detect
Daniel Golle [Sun, 4 Nov 2018 17:22:33 +0000 (18:22 +0100)]
oxnas: switch to generic board detect

remove /lib/oxnas.sh platform-specific board-detection and use
generic which is based on device-tree compatible node instead.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
6 years agoiperf: allow non-ipv6 builds
Alexander Couzens [Sat, 3 Nov 2018 00:22:59 +0000 (01:22 +0100)]
iperf: allow non-ipv6 builds

Add configure argument --disable-ipv6 when ipv6 is deselected.
Add fix-non-ipv6-builds.patch as long there is no new upstream
release.

Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
6 years agoramips: fix leds on GL.iNet GL-MT300N-V2
Martin Weinelt [Fri, 2 Nov 2018 19:52:01 +0000 (20:52 +0100)]
ramips: fix leds on GL.iNet GL-MT300N-V2

The WAN LED now shows the link state. It's color is green,
not blue.

Signed-off-by: Martin Weinelt <hexa@darmstadt.ccc.de>
6 years agocurl: noop commit to refer CVEs fixed in 7.62.0
Hans Dedecker [Fri, 2 Nov 2018 12:05:13 +0000 (13:05 +0100)]
curl: noop commit to refer CVEs fixed in 7.62.0

When bumping Curl to 7.62.0 in commit 278e4eba09 I did not include the fixed
CVEs in the commit message; this commit fixes this.

The following CVEs were fixed in 7.62.0 :

CVE-2018-16839
CVE-2018-16840
CVE-2018-16842

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agoipq-wifi: drop custom board-2.bins
Christian Lamparter [Mon, 22 Oct 2018 21:47:56 +0000 (23:47 +0200)]
ipq-wifi: drop custom board-2.bins

The BDFs for all boards were upstreamed to the ath10k-firmware
repository and linux-firmware.git.

We switched to the upstream board-2.bin, hence the files can be removed
here.

Keep the ipq-wifi package in case new boards are added. It might take
some time till board-2.bins send upstream are merged.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
6 years agoipq40xx: nbg6617: switch to upstream boardfile variant
Christian Lamparter [Mon, 22 Oct 2018 21:47:55 +0000 (23:47 +0200)]
ipq40xx: nbg6617: switch to upstream boardfile variant

The NBG6617's boardfile was merged and this device can
now switch to the upstream ath10k board-2.bin.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
6 years agoath10k-firmware: update board-2.bin for community firmwares
Christian Lamparter [Mon, 22 Oct 2018 21:47:54 +0000 (23:47 +0200)]
ath10k-firmware: update board-2.bin for community firmwares

This patch updates the board-2.bin for the default
IPQ4019, QCA9984 and QCA9888 ath10k-firmware-xyz-ct
and -ct-htt firmwares.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
6 years agoipq40xx: rt-ac58u: replace ubi auto load hack
Christian Lamparter [Mon, 22 Oct 2018 21:47:53 +0000 (23:47 +0200)]
ipq40xx: rt-ac58u: replace ubi auto load hack

This patch replaces the custom autoload quirk of the
RT-AC58U with a bootargs-append overwrite.

The vendor's u-boot doesn't leave the bootargs / cmdline alone,
so the it can't be overwritten in any other way right now...
And of course, this will be a lot of fun to deal with once
the device switches to the new spi-nand subsystem.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
6 years agokernel: Remove dependencies on old kernels
Rosen Penev [Sun, 28 Oct 2018 01:53:52 +0000 (18:53 -0700)]
kernel: Remove dependencies on old kernels

Kernels 4.1 and 4.4 are not part of the tree anymore.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
6 years agotools/patch: Add fedora patch for crashing git style patches
Rosen Penev [Wed, 31 Oct 2018 23:55:14 +0000 (16:55 -0700)]
tools/patch: Add fedora patch for crashing git style patches

https://lists.gnu.org/archive/html/bug-patch/2018-10/msg00000.html

I assume a CVE number will be assigned soon.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
6 years agolibconfig: Fix tarball filename (and URL)
Daniel Engberg [Fri, 19 Oct 2018 21:05:44 +0000 (23:05 +0200)]
libconfig: Fix tarball filename (and URL)

v$(PKG_VERSION).tar.gz is a bad idea and will clash for obvious reasons.

Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
6 years agoscripts: update config.guess and config.sub
Hauke Mehrtens [Thu, 1 Nov 2018 13:05:59 +0000 (14:05 +0100)]
scripts: update config.guess and config.sub

This updates these two files to commit 2fa97a8a0ed3 ("config.guess
(amd64:CYGWIN*:*:*, x86_64:CYGWIN*:*:*): Set master") which is the
current master of
https://git.savannah.gnu.org/gitweb/?p=config.git;a=summary

This contains updates for multiple architectures and will unbreak the
build on the x32 ABI.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agobuild: fix libressl build on x32 (amd64ilp32) host
Thorsten Glaser [Mon, 22 Oct 2018 19:00:57 +0000 (21:00 +0200)]
build: fix libressl build on x32 (amd64ilp32) host

disable use of assembly code since x32 gets misdetected as amd64

Signed-off-by: Thorsten Glaser <tg@mirbsd.org>
6 years agouboot-mxs: bump to v2018.09
Michael Heimpold [Sun, 28 Oct 2018 14:26:08 +0000 (15:26 +0100)]
uboot-mxs: bump to v2018.09

Also update the U-Boot BSP patch for I2SE Duckbill devices
and remove upstreamed patch for LibreSSL support.

Signed-off-by: Michael Heimpold <mhei@heimpold.de>
6 years agotools/ccache: update to 3.5
Hannu Nyman [Sun, 28 Oct 2018 14:49:28 +0000 (16:49 +0200)]
tools/ccache: update to 3.5

Update ccache to 3.5

Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
6 years agomwlwifi: update to version 10.3.8.0-20181029
Kabuli Chana [Mon, 29 Oct 2018 16:58:07 +0000 (10:58 -0600)]
mwlwifi: update to version 10.3.8.0-20181029

compile / test target mvebu / mamba

Signed-off-by: Kabuli Chana <newtownBuild@gmail.com>
6 years agokernel: add kmod-usb-serial-mos7840
Thomas Fuzeau [Tue, 30 Oct 2018 11:06:22 +0000 (12:06 +0100)]
kernel: add kmod-usb-serial-mos7840

Enable build of kernel module for mos7840 usb-serial devices.

Signed-off-by: Thomas Fuzeau <tfuzeau@loocla.com>
6 years agocurl: bump to 7.62.0
Hans Dedecker [Wed, 31 Oct 2018 21:58:54 +0000 (22:58 +0100)]
curl: bump to 7.62.0

Refresh patches, for changes in version 7.62.0 see https://curl.haxx.se/changes.html#7_62_0

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agoopenssl: enable OPENSSL_WITH_DEPRECATED when OpenSSL is built as a build dep
Matthias Schiffer [Tue, 30 Oct 2018 19:38:34 +0000 (20:38 +0100)]
openssl: enable OPENSSL_WITH_DEPRECATED when OpenSSL is built as a build dep

Some package (e.g. libunbound) depend on OPENSSL_WITH_DEPRECATED. In some
situations it may happen that libunbound and openssl are only pulled in as
build dependencies, but are not enabled in .config.

In such cases, the defaults of symbols like OPENSSL_WITH_DEPRECATED are
ignored (as the whole symbol depends on PACKAGE_libopenssl), and config
symbol dependencies of libunbound aren't effective either (as libunbound
is not actually enabled).

This commit works around the issue by introducing a hidden negated symbol
OPENSSL_NO_DEPRECATED, which is always disabled when PACKAGE_libopenssl is
disabled, and ensures that OpenSSL is built with deprecated APIs in this
case. A user can still manage to break the build by explicitly enabling
libopenssl and disabling OPENSSL_WITH_DEPRECATED; the interaction between
build dependencies and config symbols will require further discussion.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
6 years agodnsmasq: tighten config file permissions
Kevin Darbyshire-Bryant [Thu, 11 Oct 2018 08:46:42 +0000 (09:46 +0100)]
dnsmasq: tighten config file permissions

Install following as config files (600) perms instead of as data (644)

/usr/share/dnsmasq/dhcpbogushostname.conf
/usr/share/dnsmasq/trust-anchors.conf
/usr/share/dnsmasq/rfc6761.conf
/etc/hotplug.d/ntp/25-dnsmasqsec
/etc/config/dhcp
/etc/dnsmasq.conf

dnsmasq reads relevant config files before dropping root privilege and
running as dnsmasq:dnsmasq

ntpd runs as root so the hotplug script is still accessible

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
6 years agoar71xx: expand kernel size for ubnt targets
Koen Vandeputte [Mon, 29 Oct 2018 14:48:10 +0000 (15:48 +0100)]
ar71xx: expand kernel size for ubnt targets

Kernel 4.14 is pretty large causing a build error as the partition is too small.
Expand the kernel partition a bit to make it fit.

* ubnt-uap-pro
* ubnt-unifi-outdoor-plus

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoar71xx: disable ap91-5g due to kernel size
Koen Vandeputte [Mon, 29 Oct 2018 14:44:54 +0000 (15:44 +0100)]
ar71xx: disable ap91-5g due to kernel size

This target causes a build failure as the kernel image does not fit
into the kernel partition.

As the kernel is located behind the rootfs, it cannot be enlarged
as the boot entry location would get altered.

Disable this target for now.

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoar71xx: fix sysupgrade generation for some targets
Koen Vandeputte [Wed, 24 Oct 2018 12:44:30 +0000 (14:44 +0200)]
ar71xx: fix sysupgrade generation for some targets

These targets are already defined as having a dynamic partition
but the safeloader was not adapted for them.

This causes a build warning for the sysupgrade image being too big.

Targets:
- c58-v1
- c60-v1
- c60-v2
- TL-WR1043 v5
- TL-WR902AC v1
- TL-WR942N V1

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoar71xx: switch more targets to dynamic partitioning
Koen Vandeputte [Wed, 24 Oct 2018 12:49:07 +0000 (14:49 +0200)]
ar71xx: switch more targets to dynamic partitioning

Targets:
- TP-LINK ER355
- TP-LINK C25 V1
- TP-LINK C59 V1
- TP-LINK C7 V4
- TP-LINK C7 V5

Fixes build issues seen due to the kernel being too big

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoar71xx: switch TP-Link RE450 v1 to dynamic partitioning
Radek Dostál [Mon, 22 Oct 2018 20:34:17 +0000 (22:34 +0200)]
ar71xx: switch TP-Link RE450 v1 to dynamic partitioning

As mentioned in commit 5f24933 recent changes on ar71xx (switch to 4.14,
memory compaction, ...) cause an increase in kernel size, making it too
big for RE450.

RE450 images were not build due to the following error message:
os-image partition too big (more than 1572864 bytes): Success

Tested on RE450, device boots and was used to send this patch.

Reported-by: Enrico Mioso <mrkiko.rs@gmail.com>
Suggested-by: David Bauer <mail@david-bauer.net>
Signed-off-by: Radek Dostál <rd@radekdostal.com>
[rewrote commit msg keeping it tight + fixed SoB lines]
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoar71xx: move OCEDO Koala to generic and use dynamic partitioning
David Bauer [Thu, 18 Oct 2018 18:43:42 +0000 (20:43 +0200)]
ar71xx: move OCEDO Koala to generic and use dynamic partitioning

This changes the OCEDO Koala flash-layout to a unified firmware
partition, thus making the ar71xx-generic kernel fit in flash.

Compile and runtested on OCEDO Koala.

Signed-off-by: David Bauer <mail@david-bauer.net>
[small title reword]
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoath79: fix PQI Air Pen support
Mathias Kresin [Sun, 28 Oct 2018 10:03:49 +0000 (11:03 +0100)]
ath79: fix PQI Air Pen support

Use the correct dts filename. It has to be <soc>_<manufacturer>_<device>.

Signed-off-by: Mathias Kresin <dev@kresin.me>
6 years agoath79: add support for PQI Air Pen
YuheiOKAWA [Thu, 4 Oct 2018 16:08:51 +0000 (01:08 +0900)]
ath79: add support for PQI Air Pen

SoC: AR9330 (or AR9331 revision?)
Ethernet x1, Wireless 2.4G, uSD card slot x1.
USB Power, include Li-Po Battery.

Flash 8MiB. (factory is dual image)
1. write image is use half 4MiB size from factory.
  write initramfs image to mtd6 and mtd7.
2. change next boot setting. next is mtd6 boot.
  echo -ne "\x01" | dd of=/dev/mtdblock8 conv=notrunc bs=1 count=1 seek=0

3. rewrite sysupgrade image.
  change next boot setting. next is mtd4.
  echo -ne "\x00" | dd of=/dev/mtdblock8 conv=notrunc bs=1 count=1 seek=0

OEM mtd.
dev:    size   erasesize  name
mtd0: 00040000 00010000 "u-boot"
mtd1: 00010000 00010000 "u-boot-env"
mtd2: 00010000 00010000 "ART"
mtd3: 00010000 00010000 "NVRAM"
mtd4: 00140000 00010000 "uImage"
mtd5: 00280000 00010000 "rootfs"
mtd6: 00140000 00010000 "uImage2"
mtd7: 00280000 00010000 "rootfs2"
mtd8: 00010000 00010000 "CONF"

Signed-off-by: YuheiOKAWA <tochiro.srchack@gmail.com>
6 years agoath79: Add support for Ubiquiti LiteAP ac (LAP-120)
Roger Pueyo Centelles [Wed, 19 Sep 2018 12:17:35 +0000 (14:17 +0200)]
ath79: Add support for Ubiquiti LiteAP ac (LAP-120)

This commit adds support for the Ubiquiti LiteAP ac (LAP-120), an outdoor
5 GHz AC access point with an integrated 120° 16 dBi antenna. The device
was previously known as LiteBeam AP ac but was later rebranded.

CPU:    Atheros AR9342 SoC
RAM:    64 MB DDR2
Flash:  16 MB NOR SPI
Ports:  1 GbE port (PoE in)
WLAN:   5 GHz QCA899X (PCI)

The integrated QCA899X is a Ubiquiti branded part with modified vendor
and product id (0777:11ac9). It is very similar to the NanoStation loco
AC, except for the 2.4 GHz management radio (missing here).

Installation procedure is the same as the NanoStation [loco] AC:

1. Connect to serial header on device
2. Power on device and enter uboot console
3. Set up tftp server serving an openwrt initramfs build
4. Load initramfs build using the command tftpboot in the uboot cli
5. Boot the loaded image using the command bootm
6. Copy squashfs openwrt sysupgrade build to the booted device
7. Use mtd to write sysupgrade to partition "firmware"
8. Reboot and enjoy

Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
6 years agomt76: update to the latest version
Felix Fietkau [Sat, 27 Oct 2018 15:54:36 +0000 (17:54 +0200)]
mt76: update to the latest version

fdc63f1 mt76: fix regression in tx status handling

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agomt76: Fix dependencies for PCI support
Daniel Engberg [Wed, 10 Oct 2018 11:04:31 +0000 (13:04 +0200)]
mt76: Fix dependencies for PCI support

USB support doesn't necessarily mean that there's
PCI support available so move that to the drivers
which requires PCI support. This applies to the
sunxi platform for instance.

Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
Signed-off-by: Felix Fietkau <nbd@nbd.name> [fix mt76x0e]
6 years agomt76: update to the latest version
Felix Fietkau [Thu, 25 Oct 2018 20:48:49 +0000 (22:48 +0200)]
mt76: update to the latest version

0b8b15f mt7603: use common mmio mcu struct
d612e2e mt76: mt76x02: skip station tx status for non-sta wcid entries
d89d84b mt76: mt76x02: only override control->sta on sw-encrypted tx
9d59d2a mt76: add support for reporting tx status with skb
ed17ba2 mt7603: use common tx status handling code
14b6c59 mt76x0: use band parameter for LC calibration

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agogrub2: fix regression caused by binutils-2.31.1
Philip Prindeville [Sun, 21 Oct 2018 02:23:12 +0000 (20:23 -0600)]
grub2: fix regression caused by binutils-2.31.1

grub-efi no longer works:

grub-mkimage: error: relocation 0x4 is not implemented yet.

See:

http://git.savannah.gnu.org/cgit/grub.git/commit/util?id=842c390469e2c2e10b5aa36700324cd3bde25875

Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com>
Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>i [PKG_RELEASE increase]
6 years agomwlwifi: driver version to 10.3.8.0-20181022
Jonathan Lancett [Mon, 22 Oct 2018 14:36:23 +0000 (14:36 +0000)]
mwlwifi: driver version to 10.3.8.0-20181022

Upgrade 88W8997 firmware to 8.4.0.52.
Removed unnecessary firmware settings.
Added vendor events.
Fixed crash problem when module is removed.
Modified the code to protect tx queues.

Signed-off-by: Jonathan Lancett <j.lancett@ntlworld.com>
[tidy commit message]
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
6 years agoglibc: update to latest 2.27 commit (BZ #23562, BZ #23275)
Hans Dedecker [Tue, 23 Oct 2018 13:55:09 +0000 (15:55 +0200)]
glibc: update to latest 2.27 commit (BZ #23562, BZ #23275)

77b4b8231e signal: Use correct type for si_band in siginfo_t [BZ #23562]
5bdb6897fc Fix race in pthread_mutex_lock while promoting to PTHREAD_MUTEX_ELISION_NP [BZ #23275]

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agomt76: update to the latest version
Felix Fietkau [Tue, 23 Oct 2018 09:17:35 +0000 (11:17 +0200)]
mt76: update to the latest version

d033272 mt76: mt76x0: handle chip specific initval differences
1a2b2dc mt76: usb: fix static tracepoints
11151f4 mt76x0: antenna select corrections
dade109 mt76x0: init: simplify mt76x0_init_mac_registers
3c54612 mt76x0: pci: add missing MODULE_FIRMWARE macro
97e0269 mt76x0: mac: remove mt76x0_mac_set_ampdu_factor
5d8c4bf mt76x0: align mt76x0u and mt76x0e fw version
d4391ae mt76: move mt76x02_mac_set_short_preamble in mt76x02_mac.c
aa6ebf9 mt76: move mt76x02_init_device in mt76x02-lib module
a4a08c3 mt76: move mac beacon routines in mt76x02-lib module
6b69650 mt76: move tx beacon routines in mt76x02-lib module
8486047 mt76x0: pci: add pre_tbtt_tasklet support
acd4c99 mt76: move mt76x02_sw_scan and mt76x02_sw_scan_complete in mt76x02-lib module
2d736cf mt76: move mt76x02_get_txpower in mt76x02_util.c
aa1d6bd mt76: move mt76x02_sta_ps in mt76x02-lib module
e684281 mt76: introduce mt76x02_init_beacon_config routine
c181d97 mt76x0: pci: enable AP support
3b86d99 mt76: move mt76x02_set_tx_ackto in mt76x02-lib module
41d7ef7 mt76x0: update init vals for MT_TX_PROT registers
fe5a149 mt76: move tx protection routines in mt76x02-lib module
5bca2de mt76: move mt76x02_bss_info_changed in mt76x02-lib module
d21a835 mt76: move dfs support in mt76x02-lib module
ba9d424 mt76x0: pci: add DFS support

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agokernel: add missing symbol for target bcm53xx
Koen Vandeputte [Tue, 23 Oct 2018 08:55:56 +0000 (10:55 +0200)]
kernel: add missing symbol for target bcm53xx

Fixes: ca88f4153f8f ("kernel: bump 4.14 to 4.14.77")
Reported-by: Rafał Miłecki <rafal@milecki.pl>
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agobcm53xx: use upstream SPI controller fix
Rafał Miłecki [Tue, 23 Oct 2018 07:42:00 +0000 (09:42 +0200)]
bcm53xx: use upstream SPI controller fix

This just moves patch to use 0xx prefix and includes maintainer's s-o-b.

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
6 years agokernel: bump 4.14 to 4.14.78
Koen Vandeputte [Mon, 22 Oct 2018 12:11:59 +0000 (14:11 +0200)]
kernel: bump 4.14 to 4.14.78

Refreshed all patches.

Remove upstreamed:
- 050-net-emac-fix-fixed-link-setup-for-the-RTL8363SB-swit.patch

Compile-tested on: cns3xxx, imx6
Runtime-tested on: cns3xxx, imx6

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agokernel: bump 4.9 to 4.9.135
Koen Vandeputte [Mon, 22 Oct 2018 09:46:59 +0000 (11:46 +0200)]
kernel: bump 4.9 to 4.9.135

Refreshed all patches.

Fixes:
- CVE-2018-10883

Compile-tested on: ar71xx, layerscape
Runtime-tested on: ar71xx

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agofstools: update to git HEAD
Hans Dedecker [Sun, 21 Oct 2018 18:43:46 +0000 (20:43 +0200)]
fstools: update to git HEAD

29e53af fstools: add ntfs support

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agomac80211: hermes-pcmcia: add missing dependency to kmod-pcmcia-core
Hauke Mehrtens [Sun, 21 Oct 2018 14:35:49 +0000 (16:35 +0200)]
mac80211: hermes-pcmcia: add missing dependency to kmod-pcmcia-core

This fixes a build problem on x86.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agolinux-firmware: intel: Add iwl 9000 and 9260 firmware
Hauke Mehrtens [Thu, 18 Oct 2018 19:03:16 +0000 (21:03 +0200)]
linux-firmware: intel: Add iwl 9000 and 9260 firmware

Add the most recent supported firmware file for the Intel 9000 and
9260 wireless chips. The API version 41 is not yet supported by the driver.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agolinux-firmware: intel: update firmware files for 7265D, 8000C and 8265
Hauke Mehrtens [Thu, 18 Oct 2018 19:02:20 +0000 (21:02 +0200)]
linux-firmware: intel: update firmware files for 7265D, 8000C and 8265

Use more recent versions for the Intel wireless chips 7265D, 8000C and 8265.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
6 years agompc85xx: workaround 4.14 cross compile issue
Ryan Mounce [Fri, 12 Oct 2018 07:46:28 +0000 (18:16 +1030)]
mpc85xx: workaround 4.14 cross compile issue

Regression introduced with 4.10 by 43c9127d94d62a232ed33ed2eab8a08657ce5472

Build will fail if system 'ar' does not support deterministic builds.
e.g. macOS with Xcode toolchain

Appears to be fixed upstream in 4.18 by af3901cbbd3de182aafb8ee553c825c0074df6a2

Signed-off-by: Ryan Mounce <ryan@mounce.com.au>
6 years agotools/pkg-config: pass arguments at the end
Arthur Skowronek [Mon, 15 Oct 2018 12:55:57 +0000 (14:55 +0200)]
tools/pkg-config: pass arguments at the end

Go for openwrt passes pkg-config arguments in the format of

        pkg-config --cflags -- pkg-name

which in turn will be passed down to the real pkg-config as something
like

        pkg-config.real --cflags -- pkg-name --define...

and causes the real pkg-config implementation to missinterpret the given
argument list.

This also helps to fix https://github.com/golang/go/issues/27940

Signed-off-by: Arthur Skowronek <arthur.skowronek@tuta.io>
6 years agomwlwifi: driver version to 10.3.8.0-20181008
Yufei Miao [Wed, 10 Oct 2018 15:23:44 +0000 (23:23 +0800)]
mwlwifi: driver version to 10.3.8.0-20181008

Signed-off-by: Yufei Miao <myf@myf.cloud>
6 years agokernel/modules: Aosong AM2315/AM2320 IIO kernel module support enabled.
Roman Bazalevskiy [Fri, 12 Oct 2018 12:16:42 +0000 (15:16 +0300)]
kernel/modules: Aosong AM2315/AM2320 IIO kernel module support enabled.

Signed-off-by: Roman Bazalevskiy <rvb@rvb.name>
6 years agokernel/modules: add kmod-fou6 package
Deng Qingfang [Tue, 16 Oct 2018 15:01:05 +0000 (23:01 +0800)]
kernel/modules: add kmod-fou6 package

IPv6 support for Foo-over-UDP tunnel

Signed-off-by: Deng Qingfang <dengqf6@mail2.sysu.edu.cn>
6 years agolinux-firmware: Update linux-firmware to 2018-10-18
Robert Marko [Fri, 19 Oct 2018 07:54:32 +0000 (09:54 +0200)]
linux-firmware: Update linux-firmware to 2018-10-18

Kalles ath10k PR was finally merged so update linux-firmware to
include those changes.

This is needed since disabling ath10k-firmware a lot of custom BDF-s
in board-2.bin-s are not available in previously outdated linux-firmware
board-2.bin-s.
This also includes support for boards currently using ipq-wifi and other
WIP ones.

Runtime tested on 8devices Jalapeno.

Signed-off-by: Robert Marko <robimarko@gmail.com>
6 years agoapm821xx: add disk-activity triggers via DT.
Christian Lamparter [Mon, 15 Oct 2018 18:00:09 +0000 (20:00 +0200)]
apm821xx: add disk-activity triggers via DT.

The WNDR4700 and the MBL have a dedicated HDD activity LED.
This patch adds the default led triggers to the DTS and
removes the entries in 01_leds.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
6 years agoapm821xx: apm82181.dtsi overhaul
Christian Lamparter [Mon, 15 Oct 2018 18:00:08 +0000 (20:00 +0200)]
apm821xx: apm82181.dtsi overhaul

- order entries by mmio address where possible

 - switch to lower-case address values

 - comment on BSP u-boot behaviour in regards to
   what it edits and look for

 - annotate irq lines with the help of interrupt-names

 - remove deprecated "device_type" properties
   The pci and network device_type had to stay since
   they are required by the drivers and u-boot.
   the cpu and memory device_types will remain as well
   as they are still part of the DT Spec.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
6 years agoapm821xx: force WNDR4700 console to 115200n8
Christian Lamparter [Mon, 15 Oct 2018 18:00:07 +0000 (20:00 +0200)]
apm821xx: force WNDR4700 console to 115200n8

With the upcoming 4.19 release, the serial console on the
Netgear WNDR4700 would no longer work as it is never really
set and always relied on the kernel's serial code not to
change the baud rate.

On the stock firmware, Netgear forced the console setting
through a custom CONFIG_CMDLINE in their kernel to 115200.
Normally, they should have done it in a different way and
just passed the baudrate through a "console=..." kernel
parameter via the bootargs in the "/chosen" dt node.
But in their default u-boot bootcmd setting, they somehow
forgot to add the "run addtty", so there's no easy way to
pass the baudrate from u-boot to the kernel. So it has
to be forced as otherwise it ends up as 9600 baud.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
6 years agoapm821xx: set uart compatible to ns16750
Christian Lamparter [Mon, 15 Oct 2018 18:00:06 +0000 (20:00 +0200)]
apm821xx: set uart compatible to ns16750

The APM821XX/PPC460EX datasheet mentions:
"Register compatibility with 16750 register set"

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
6 years agobuild: remove obsolete -rc kernel testing rewrites
Christian Lamparter [Mon, 15 Oct 2018 18:00:05 +0000 (20:00 +0200)]
build: remove obsolete -rc kernel testing rewrites

The -rcX "testing" kernels are no longer hosted on
cdn.kernel.org file servers directly in a "testing"
directory. Therefore the logic that tested for "-rc"
can be removed.

Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
6 years agogemini: Fix up image generation
Linus Walleij [Wed, 17 Oct 2018 18:46:19 +0000 (20:46 +0200)]
gemini: Fix up image generation

I noticed that the image generator for the Gemini generates
some 10+ MB files for the second (application) partition.
This is just wrong. The first 6144K partition named initramfs
easily fits OpenWRTs squashfs,jffs2 overlayed partitions
with nice headroom for storing configuration files.

Generate a blank partition for the hddapp.tgz file and delete
it after generating the firmware image - when performing
updates manually you just want the "rd.z" file around
anyways.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
6 years agogemini: Break out USB to packages
Linus Walleij [Thu, 18 Oct 2018 09:30:27 +0000 (11:30 +0200)]
gemini: Break out USB to packages

This creates a new kernel package for the fotg210 host
controller and uses that with the gemini to shrink the
kernel. The SQ201 needs the USB2 PCI package as well.

The build system required me to make kernel_oldconfig
beofore it would build without errors so some minor
unrelated Kconfig entries are changed.

Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
6 years agokernel: Add support for Winbond w25q128jv SPI NOR flash
Baptiste Jonglez [Thu, 18 Oct 2018 09:07:26 +0000 (11:07 +0200)]
kernel: Add support for Winbond w25q128jv SPI NOR flash

Newer batches of several Mikrotik boards contain this yet-unsupported
flash chip, for instance:

- rb941-2nd (hAP lite)
- rb952ui-5ac2nd (hAP ac lite)
- RBM33G

and probably other Mikrotik boards need this patch as well.

The patch was submitted upstream by Robert Marko: https://patchwork.ozlabs.org/patch/934181/

Closes: FS#1715
Signed-off-by: Baptiste Jonglez <git@bitsofnetworks.org>
Cc: Robert Marko <robimarko@gmail.com>
6 years agotoolchain/musl: Don't override optimization set by upstream
Daniel Engberg [Thu, 22 Feb 2018 22:39:40 +0000 (23:39 +0100)]
toolchain/musl: Don't override optimization set by upstream

Don't override optimization set by upstream.
Provides a speed increase for internal (library), malloc and string
operations in musl.

Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
[add : to PKG_RELEASE release variable for consistency]
Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
6 years agodnsmasq: bump to v2.80
Kevin Darbyshire-Bryant [Fri, 19 Oct 2018 14:17:23 +0000 (15:17 +0100)]
dnsmasq: bump to v2.80

dnsmasq v2.80 release

Change from rc1:

91421cb Fix compiler warning.

Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
6 years agokernel: bump 4.14 to 4.14.77
Koen Vandeputte [Thu, 18 Oct 2018 10:09:22 +0000 (12:09 +0200)]
kernel: bump 4.14 to 4.14.77

Refreshed all patches.

Altered patches:
- 666-Add-support-for-MAP-E-FMRs-mesh-mode.patch

New symbol for arm targets:
- HARDEN_BRANCH_PREDICTOR

Compile-tested on: ar71xx, cns3xxx, imx6
Runtime-tested on: ar71xx, cns3xxx, imx6

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agokernel: bump 4.9 to 4.9.134
Koen Vandeputte [Thu, 18 Oct 2018 09:58:32 +0000 (11:58 +0200)]
kernel: bump 4.9 to 4.9.134

Refreshed all patches.

Compile-tested on: ar71xx, layerscape
Runtime-tested on: ar71xx

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
6 years agoiproute2: install ip-tiny and ip-full in /usr/libexec
Hans Dedecker [Tue, 16 Oct 2018 12:16:48 +0000 (14:16 +0200)]
iproute2: install ip-tiny and ip-full in /usr/libexec

Install the ip-tiny and ip-full variants in /usr/libexec as the suffixed
ip variants are not meant to be called directly

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
6 years agomt76: update to the latest version
Felix Fietkau [Thu, 18 Oct 2018 14:27:46 +0000 (16:27 +0200)]
mt76: update to the latest version

b51337a mt76: clean up unused leftover EXPORT_SYMBOLs
383b228 mt76x0: reduce duplication in setting rf bandwidth parameters
a1a4528 mt76: mt76x0: handle chip specific initval differences
f473340 mt76: usb: fix static tracepoints
3a975b4 mt76x0: antenna select corrections
37c1150 mt76x0: init: simplify mt76x0_init_mac_registers

Signed-off-by: Felix Fietkau <nbd@nbd.name>
6 years agowireguard: bump to 0.0.20181018
Jason A. Donenfeld [Thu, 18 Oct 2018 01:48:34 +0000 (03:48 +0200)]
wireguard: bump to 0.0.20181018

ba2ab5d version: bump snapshot
5f59c76 tools: wg-quick: wait for interface to disappear on freebsd
ac7e7a3 tools: don't fail if a netlink interface dump is inconsistent
8432585 main: get rid of unloaded debug message
139e57c tools: compile on gnu99
d65817c tools: use libc's endianness macro if no compiler macro
f985de2 global: give if statements brackets and other cleanups
b3a5d8a main: change module description
296d505 device: use textual error labels always
8bde328 allowedips: swap endianness early on
a650d49 timers: avoid using control statements in macro
db4dd93 allowedips: remove control statement from macro by rewriting
780a597 global: more nits
06b1236 global: rename struct wireguard_ to struct wg_
205dd46 netlink: do not stuff index into nla type
2c6b57b qemu: kill after 20 minutes
6f2953d compat: look in Kbuild and Makefile since they differ based on arch
a93d7e4 create-patch: blacklist instead of whitelist
8d53657 global: prefix functions used in callbacks with wg_
123f85c compat: don't output for grep errors

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
6 years agoar71xx/ath79: fix mtd corruption
Fabio Bettoni [Wed, 17 Oct 2018 12:21:41 +0000 (14:21 +0200)]
ar71xx/ath79: fix mtd corruption

In commit fec8fe806963 ("kernel: bump 4.9 to 4.9.116") [1], the following patch for removed:
- 403-mtd_fix_cfi_cmdset_0002_status_check.patch

This patch contained fixes for both write and erase functions.

While the chip-detects for erase got fixed upstream [2],
some modifications are still required, even with the fixes applied.

While at it, also apply the same fix for target ath79,
which suffers the same issue.

Not doing so results in following errors seen:

Collected errors:
 * pkg_write_filelist: Failed to open //usr/lib/opkg/info/luci-lib-ip.list: I/O error.
 * opkg_install_pkg: Failed to extract data files for luci-lib-ip. Package debris may remain!
 * opkg_install_cmd: Cannot install package luci-ssl.
 * opkg_conf_write_status_files: Can't open status file //usr/lib/opkg/status: I/O error.

[    0.780920] jffs2: version 2.2 (NAND) (SUMMARY) (LZMA) (RTIME) (CMODE_PRIORITY) (c) 2001-2006 Red Hat, Inc.
[    8.406396] jffs2: notice: (415) jffs2_build_xattr_subsystem: complete building xattr subsystem, 0 of xdatum (0 unchecked, 0 orphan) and 0 of xref (0 dead, 0 orphan) found.
[    8.423476] mount_root: switching to jffs2 overlay
[  270.902671] jffs2: Write of 1989 bytes at 0x005ce6f8 failed. returned -5, retlen 962
[  270.931965] jffs2: Write of 1989 bytes at 0x005ceec0 failed. returned -5, retlen 0
[  270.939631] jffs2: Not marking the space at 0x005ceec0 as dirty because the flash driver returned retlen zero
[  270.950397] jffs2: Write of 68 bytes at 0x005ceec0 failed. returned -5, retlen 0
[  270.957838] jffs2: Not marking the space at 0x005ceec0 as dirty because the flash driver returned retlen zero
[  270.968584] jffs2: Write of 68 bytes at 0x005ceec0 failed. returned -5, retlen 0
[  270.976027] jffs2: Not marking the space at 0x005ceec0 as dirty because the flash driver returned retlen zero
[  270.986735] jffs2: Write of 68 bytes at 0x005ceec0 failed. returned -5, retlen 0
[  270.994225] jffs2: Not marking the space at 0x005ceec0 as dirty because the flash driver returned retlen zero

[1] https://git.openwrt.org/?p=openwrt/openwrt.git;a=commit;h=fec8fe806963c96a6506c2aebc3572d3a11f285f
[2] https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v4.9.133&id=a0239d83e1cb60de5e78452d4708c083b9e3dcbe

Fixes: fec8fe806963 ("kernel: bump 4.9 to 4.9.116")
Signed-off-by: Fabio Bettoni <fbettoni@gmail.com>
Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>