openwrt/staging/robimarko.git
5 years agoramips: add support for XiaoYu XY-C5
Qi Jiang [Sun, 8 Sep 2019 01:52:07 +0000 (09:52 +0800)]
ramips: add support for XiaoYu XY-C5

Hardware:
SoC:      MT7621A
Flash:    32 MiB
RAM:      512 MiB
Ethernet: built-in switch
USB:      1x USB3.0
SATA:     ASM1060, 1 SATA port

Flash instruction (TFTP):
1. Set PC to fixed ip address 192.168.173.22
2. Download *-sysupgrade.bin image and rename it to firmware.img
3. Start a tftp server with the image file in its root directory
4. Turn off the router
5. Turn on router,press the reset button and wait ~15 seconds
6. Release the reset button and after a short time
   the firmware should be transferred from the tftp server
7. Wait ~3 minutes to complete recovery.

Signed-off-by: Qi Jiang <rushx@live.cn>
[squash commits, add label-mac-device, fix sign-off style]
Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
5 years agokernel: bump 4.19 to 4.19.75
Koen Vandeputte [Tue, 24 Sep 2019 09:05:57 +0000 (11:05 +0200)]
kernel: bump 4.19 to 4.19.75

Refreshed all patches.

Fixes:
- CVE-2019-14814
- CVE-2019-14815
- CVE-2019-14816
- CVE-2019-14821

Compile-tested on: cns3xxx
Runtime-tested on: cns3xxx

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
5 years agokernel: bump 4.14 to 4.14.146
Koen Vandeputte [Tue, 24 Sep 2019 09:05:28 +0000 (11:05 +0200)]
kernel: bump 4.14 to 4.14.146

Refreshed all patches.

Fixes:
- CVE-2019-14814
- CVE-2019-14815
- CVE-2019-14816
- CVE-2019-14821

Compile-tested on: cns3xxx
Runtime-tested on: cns3xxx

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
5 years agokernel: bump 4.9 to 4.9.194
Koen Vandeputte [Tue, 24 Sep 2019 09:00:47 +0000 (11:00 +0200)]
kernel: bump 4.9 to 4.9.194

Refreshed all patches.

Fixes:
- CVE-2019-14814
- CVE-2019-14815
- CVE-2019-14816
- CVE-2019-14821

Compile-tested on: none
Runtime-tested on: none

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
5 years agoiwinfo: update to latest Git HEAD
Jo-Philipp Wich [Sun, 22 Sep 2019 16:57:58 +0000 (18:57 +0200)]
iwinfo: update to latest Git HEAD

313e827 nl80211: keep awaiting wpa_supplicant scan results on busy response
a766751 nl80211: fix parsing of mixed wpa encryption in wpa_supp scan results
f096bfd utils: support parsing SAE and OWE key management suites from IEs

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
5 years agoath79: convert devices to interrupt-driven gpio-keys
Adrian Schmutzler [Sat, 10 Aug 2019 17:22:40 +0000 (19:22 +0200)]
ath79: convert devices to interrupt-driven gpio-keys

This converts all remaining devices to use interrupt-driven
gpio-keys compatible instead of gpio-keys-polled.
The poll-interval is removed.

Only ar7240_netgear_wnr612-v2 is kept at gpio-keys-polled, as
this one is using ath9k keys.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Tested-by: Karl Palsson <karlp@etactica.com>
Tested-by: Dmitry Tunin <hanipouspilot@gmail.com>
5 years agohostapd: Add mesh support for wpad full
Hauke Mehrtens [Sat, 7 Sep 2019 14:17:14 +0000 (16:17 +0200)]
hostapd: Add mesh support for wpad full

This increases the size of the binary slightly:

old:
427722 wpad-wolfssl_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk
431696 wpad-openssl_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk

new:
442109 wpad-wolfssl_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk
445997 wpad-openssl_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
5 years agohostapd: use getrandom syscall
Hauke Mehrtens [Sat, 4 May 2019 12:02:07 +0000 (14:02 +0200)]
hostapd: use getrandom syscall

hostapd will not use the getrandom() syscall and as a fallback use
/dev/random, the syscall is supported since Linux 3.17 and in the musl,
glibc and uclibc version used by OpenWrt.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
5 years agohostapd: Remove unneeded patch
Hauke Mehrtens [Sat, 4 May 2019 00:01:15 +0000 (02:01 +0200)]
hostapd: Remove unneeded patch

All the content of this function is proceeded by IEEE8021X_EAPOL no code
accesses the ssid variable outside of this ifdef.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
5 years agohostapd: use config option CONFIG_NO_LINUX_PACKET_SOCKET_WAR
Hauke Mehrtens [Fri, 3 May 2019 23:58:53 +0000 (01:58 +0200)]
hostapd: use config option CONFIG_NO_LINUX_PACKET_SOCKET_WAR

Instead of patching the workaround away, just use the config option.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
5 years agohostapd: Update to version 2.9 (2019-08-08)
Hauke Mehrtens [Sat, 10 Aug 2019 14:49:34 +0000 (16:49 +0200)]
hostapd: Update to version 2.9 (2019-08-08)

The size of the ipkgs increase a bit (between 0.7% and 1.1%):

old 2019-04-21 (2.8):
288264 wpad-basic_2019-04-21-63962824-1_mipsel_24kc.ipk
256188 wpad-mini_2019-04-21-63962824-1_mipsel_24kc.ipk
427475 wpad-openssl_2019-04-21-63962824-1_mipsel_24kc.ipk
423071 wpad-wolfssl_2019-04-21-63962824-1_mipsel_24kc.ipk

new 2019-08-08 (2.9):
290217 wpad-basic_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk
258745 wpad-mini_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk
431732 wpad-openssl_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk
427641 wpad-wolfssl_2019-08-08-ca8c2bd2-1_mipsel_24kc.ipk

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
5 years agohostapd: Update to version 2.8 (2019-04-21)
Hauke Mehrtens [Fri, 3 May 2019 23:52:25 +0000 (01:52 +0200)]
hostapd: Update to version 2.8 (2019-04-21)

This also syncs the configuration files with the default configuration
files, but no extra options are activated or deactivated.

The mesh patches were partially merged into hostapd 2.8, the remaining
patches were extracted from patchwork and are now applied by OpenWrt.
The patches still have open questions which are not fixed by the author.
They were taken from this page:
https://patchwork.ozlabs.org/project/hostap/list/?series=62725&state=*

The changes in 007-mesh-apply-channel-attributes-before-running-Mesh.patch
where first applied to hostapd, but later reverted in hostapd commit
3e949655ccc5 because they caused memory leaks.

The size of the ipkgs increase a bit (between 1.3% and 2.3%):

old 2018-12-02 (2.7):
283337 wpad-basic_2018-12-02-c2c6c01b-11_mipsel_24kc.ipk
252857 wpad-mini_2018-12-02-c2c6c01b-11_mipsel_24kc.ipk
417473 wpad-openssl_2018-12-02-c2c6c01b-11_mipsel_24kc.ipk
415105 wpad-wolfssl_2018-12-02-c2c6c01b-11_mipsel_24kc.ipk

new 2019-04-21 (2.8):
288264 wpad-basic_2019-04-21-63962824-1_mipsel_24kc.ipk
256188 wpad-mini_2019-04-21-63962824-1_mipsel_24kc.ipk
427475 wpad-openssl_2019-04-21-63962824-1_mipsel_24kc.ipk
423071 wpad-wolfssl_2019-04-21-63962824-1_mipsel_24kc.ipk

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Tested-by: Stefan Lippers-Hollmann <s.l-h@gmx.de>
5 years agoapm821xx: remove unnecessary kernel padding
Russell Senior [Sat, 21 Sep 2019 08:58:07 +0000 (01:58 -0700)]
apm821xx: remove unnecessary kernel padding

The allocation of LEBs to ubi volumes is handled by the sysupgrade script:

  package/base-files/files/lib/upgrade/nand.sh

and the ubimkvol and or ubirsvol command. Therefore, padding of the
kernel blob is not needed at all, so use cat instead of dd. The
BLOCKSIZE variable was only used in the dd command.  In any case, 63k
made no sense for the way BLOCKSIZE was being used.

63k (64512) does make sense for DTB_SIZE because of the offsets expected
by u-boot given extant u-boot-env variables.

Tested on Meraki MR24.

Signed-off-by: Russell Senior <russell@personaltelco.net>
5 years agoath79: add support of Netgear WNDR3800CH
Dmitry Tunin [Wed, 7 Aug 2019 20:07:33 +0000 (23:07 +0300)]
ath79: add support of Netgear WNDR3800CH

Add support for the ar71xx supported Netgear WNDR3800CH to ath79.
The device is identical to WNDR3800 except NETGEAR_BOARD_ID.

Signed-off-by: Dmitry Tunin <hanipouspilot@gmail.com>
Reviewed-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agoath79: DIR825B1 switch to gpio-keys
Dmitry Tunin [Fri, 30 Aug 2019 19:17:24 +0000 (22:17 +0300)]
ath79: DIR825B1 switch to gpio-keys

Signed-off-by: Dmitry Tunin <hanipouspilot@gmail.com>
5 years agoipq40xx: fix hw-crypto detection of qce driver
Eneas U de Queiroz [Fri, 20 Sep 2019 20:48:04 +0000 (17:48 -0300)]
ipq40xx: fix hw-crypto detection of qce driver

This adds the CRYPTO_ALG_KERN_DRIVER_ONLY flag to Qualcomm crypto engine
driver algorithms, so that openssl devcrypto can recognize them as
hardware-accelerated.

Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
[refresh, move to ipq40xx as its the only target right now]
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
5 years agoath79: remove invalid uses of ath9k_patch_fw_mac_crc
Adrian Schmutzler [Wed, 18 Sep 2019 15:11:06 +0000 (17:11 +0200)]
ath79: remove invalid uses of ath9k_patch_fw_mac_crc

Some ar9344-based devices are using ath9k_patch_fw_mac_crc, which
is meant to generate a checksum, for fixing their ath9k MAC
addresses.
However, those do not have a checksum field, and the calculated
checksum offset would be negative.

This patch will use ath9k_patch_fw_mac function for those devices.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agoramips: replace backticks by $(...)
Adrian Schmutzler [Fri, 20 Sep 2019 12:07:56 +0000 (14:07 +0200)]
ramips: replace backticks by $(...)

This replaces deprecated backticks by more versatile $(...) syntax.

While at it, remove some useless cat commands and deprecated
egrep commands.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agoath79: image: pad kernel for Adtran/Bluesocket devices
Tomasz Maciej Nowak [Tue, 17 Sep 2019 14:21:51 +0000 (16:21 +0200)]
ath79: image: pad kernel for Adtran/Bluesocket devices

It has been reported that using the sysupgrade-tar image will trigger
"lzma_decode failed error". The RedBoot bootloader always loads data
from flash till block size boundary, so if there's no padding it'll also
load the beginning of rootfs, and it seems that lzma_decoder can't handle
that garbage data. Previously the script creating combined-image
silently padded the kernel and rootfs, but since sysupgrade-tar doesn't,
pad the default kernel image.

Fixes: 900330f ("ath79: image: retire combined-image for Adtran/Bluesocket
devices")
Cc: Brian Gonyer <bgonyer@gmail.com>
Cc: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl>
5 years agoath79: sort device definitions in image/*
Adrian Schmutzler [Thu, 22 Aug 2019 12:34:45 +0000 (14:34 +0200)]
ath79: sort device definitions in image/*

This applies alphabetic sorting to devices in image/* files.

For certain cases, this patch deviates from strict sorting, e.g.
to ensure that v10 comes after v9.

While at it, fix an indent and remove some useless empty lines.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agoltq-vdsl-fw: update firmware filename and download URL
Daniel Golle [Sat, 21 Sep 2019 13:09:55 +0000 (15:09 +0200)]
ltq-vdsl-fw: update firmware filename and download URL

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
5 years agorpcd: update to latest Git HEAD
Jo-Philipp Wich [Sat, 21 Sep 2019 13:03:48 +0000 (15:03 +0200)]
rpcd: update to latest Git HEAD

95f0973 file: increase minimum read buffer size to 4096 bytes

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
5 years agobuild: set TARGET_ROOTFS_PARTSIZE to make combined image fit in 128MB
Matthias Schiffer [Sat, 21 Sep 2019 11:21:36 +0000 (13:21 +0200)]
build: set TARGET_ROOTFS_PARTSIZE to make combined image fit in 128MB

Change TARGET_ROOTFS_PARTSIZE from 128 to 104 MiB, so the whole image
(bootloader + boot + root) will fit on a 128MB CF card by default.

With these settings, the generated images (tested on x86-generic and
x86-64) have 126,353,408 bytes; the smallest CF card marketed as "128MB"
that I found a datasheet for (a Transcend TS128MCF80) has 126,959,616
bytes.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
5 years agorootfs.mk: ensure all timestamp are set to SOURCE_DATE_EPOCH
Alexander Couzens [Fri, 20 Sep 2019 03:36:13 +0000 (05:36 +0200)]
rootfs.mk: ensure all timestamp are set to SOURCE_DATE_EPOCH

Some tools doesn't support SOURCE_DATE_EPOCH (e.g. initramfs images).
Ensure all files of a root filesystem are set to SOURCE_DATE_EPOCH.
Make initramfs builds reproducible (for ramips).

Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
5 years agousign: update to latest Git HEAD
Hauke Mehrtens [Fri, 20 Sep 2019 23:05:42 +0000 (01:05 +0200)]
usign: update to latest Git HEAD

f34a383 main: fix some resource leaks

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
5 years agofstools: update to latest Git HEAD
Hauke Mehrtens [Fri, 20 Sep 2019 23:02:11 +0000 (01:02 +0200)]
fstools: update to latest Git HEAD

4327ed4 mkdev: Avoid out of bounds read
9b3eb63 libblkid-tiny: use blkid_probe_set_utf8label for label set
c9d0462 libblkid-tiny: adds blkid_probe_set_utf8label support

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
5 years agoprocd: update to the latest git HEAD
Hauke Mehrtens [Fri, 20 Sep 2019 22:57:55 +0000 (00:57 +0200)]
procd: update to the latest git HEAD

8e9fb51 procd: Switch to nanosleep
c844ace system: Fix possible integer overflows

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
5 years agohostapd: Fix AP mode PMF disconnection protection bypass
Hauke Mehrtens [Fri, 20 Sep 2019 20:20:16 +0000 (22:20 +0200)]
hostapd: Fix AP mode PMF disconnection protection bypass

This fixes
* CVE-2019-16275 AP mode PMF disconnection protection bypass
https://w1.fi/security/2019-7/ap-mode-pmf-disconnection-protection-bypass.txt

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
5 years agosunxi: Add support for Banana Pi M2 Ultra
Hal Martin [Sat, 7 Sep 2019 16:28:18 +0000 (16:28 +0000)]
sunxi: Add support for Banana Pi M2 Ultra

CPU: Allwinner R40 Quad-Core Cortex-A7 @ 1.2GHz
GPU: Mali 400 MP2
Memory: 2GB DDR3 onboard (shared with GPU)
Onboard: Storage microSD card slot
Onboard: Storage 8GB eMMC
Onboard: Network 10/100/1000M Ethernet RJ45
Onboard: Network WiFi 802.11b/g/n 1x1 (AMPAK AP6212; brcmfmac)
Onboard header: SPI, I2C, GPIO, UART
USB: Two USB 2.0 Host, One USB 2.0 OTG

Known issues:
- WiFi doesn't work
- eMMC not supported

Signed-off-by: Hal Martin <hal.martin@gmail.com>
5 years agoelfutils: Add host build
Rosen Penev [Thu, 5 Sep 2019 23:57:39 +0000 (16:57 -0700)]
elfutils: Add host build

Needed for glib2 host build:

gresource-tool.c:32:20: fatal error: libelf.h: No such file or directory
 #include <libelf.h>

Changed PKG_LICENSE to the SPDX version.

Switched build dependency for argp-standalone to !USE_GLIBC. argp is a
glibc extension. Treat it as such.

Adjusted patch to use strerror_l, which works properly with both glibc
and musl. The patch errors under glibc with:

dwfl_error.c:158:7: error: ignoring return value of 'strerror_r', declared
with attribute warn_unused_result [-Werror=unused-result]
       strerror_r (error & 0xffff, s, sizeof(s));

void casting does not fix the error.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
5 years agoat91:sam9x: create sdcard images
Sandeep Sheriker M [Wed, 11 Sep 2019 18:35:56 +0000 (18:35 +0000)]
at91:sam9x: create sdcard images

create sdcard images for sam9x25 & sam9x35 ek targets.

Signed-off-by: Sandeep Sheriker M <sandeep.sheriker@microchip.com>
5 years agoat91:sama5: create fit image
Sandeep Sheriker M [Wed, 11 Sep 2019 18:35:55 +0000 (18:35 +0000)]
at91:sama5: create fit image

create fit image & add it to sdcard image for sama5 target
device.

Signed-off-by: Sandeep Sheriker M <sandeep.sheriker@microchip.com>
5 years agowolfssl: allow building with hw-crytpo and AES-CCM
Eneas U de Queiroz [Thu, 12 Sep 2019 20:00:00 +0000 (17:00 -0300)]
wolfssl: allow building with hw-crytpo and AES-CCM

Hardware acceleration was disabled when AES-CCM was selected as a
workaround for a build failure.  This applies a couple of upstream
patches fixing this.

Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
5 years agombedtls: update to 2.16.3
Magnus Kroken [Wed, 18 Sep 2019 19:14:18 +0000 (21:14 +0200)]
mbedtls: update to 2.16.3

Remove 300-bn_mul.h-Use-optimized-MULADDC-code-only-on-ARM-6.patch,
the issue has been fixed upstream.

Signed-off-by: Magnus Kroken <mkroken@gmail.com>
5 years agouClibc++: Remove faulty patch
Rosen Penev [Fri, 20 Sep 2019 00:39:15 +0000 (17:39 -0700)]
uClibc++: Remove faulty patch

This patch was originally added to fix compilation with v4l2rtspserver.
Turns out it was v4l2rtspserver that was broken, not uClibc++. This now
causes issues with a different package where the arguments are being
split.

Note that with this patch, shellcheck throws an error:

SC2068: Double quote array expansions to avoid re-splitting elements.

More: https://github.com/openwrt/packages/pull/9972#discussion_r324878373

Signed-off-by: Rosen Penev <rosenp@gmail.com>
5 years agogcc: Fix ICE in GCC 9.2.0
Hauke Mehrtens [Wed, 18 Sep 2019 16:12:29 +0000 (18:12 +0200)]
gcc: Fix ICE in GCC 9.2.0

This backports a fix from GCC master to fix a internal compiler
exception seen when compiling libjson-c with mips16 activated.

Fixes: FS#2455
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
5 years agokernel: bump 4.19 to 4.19.74
Koen Vandeputte [Fri, 20 Sep 2019 13:11:19 +0000 (15:11 +0200)]
kernel: bump 4.19 to 4.19.74

Refreshed all patches.

Fixes:
- CVE-2019-15504

Compile-tested on: cns3xxx
Runtime-tested on: cns3xxx

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
5 years agokernel: bump 4.14 to 4.14.145
Koen Vandeputte [Fri, 20 Sep 2019 13:08:16 +0000 (15:08 +0200)]
kernel: bump 4.14 to 4.14.145

Refreshed all patches.

Compile-tested on: cns3xxx
Runtime-tested on: cns3xxx

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
5 years agoiwinfo: update to latest Git HEAD
Jo-Philipp Wich [Fri, 20 Sep 2019 11:32:49 +0000 (13:32 +0200)]
iwinfo: update to latest Git HEAD

02112f9 cli: fix reporting of mixed WPA2/WPA3 versions
7faeaea nl80211: properly detect WEP encryption in wpa_supp scan results
629b5ff nl80211: do not confuse open connections with WEP ones
3d47ddd nl80211: rework hostapd and wpa_supplicant wpa suite parsing

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
5 years agohostapd: mirror ieee80211w ap mode defaults in station mode
Jo-Philipp Wich [Fri, 20 Sep 2019 11:20:21 +0000 (13:20 +0200)]
hostapd: mirror ieee80211w ap mode defaults in station mode

For AP mode, OpenWrt automatically sets ieee80211w to either 1 or 2, depending
on whether the encryption is set to sae-mixed, or sae/owe/eap suite-b.

Mirror the same defaults for client mode connections, in order to allow an
OpenWrt station to associate to an OpenWrt ap with SAE, OWE or Suite-B encryption
without the need to manually specify "option ieee80211w" on the station.

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
5 years agohostapd: fix OWE settings in client mode
Jo-Philipp Wich [Fri, 20 Sep 2019 09:40:52 +0000 (11:40 +0200)]
hostapd: fix OWE settings in client mode

This changes fixes the generation of the wpa_supplicant client configuration
in WPA3 OWE client mode. Instead of incorrectly emitting key_mgmt=NONE, use
the proper key_mgmt=OWE setting instead.

Signed-off-by: Jo-Philipp Wich <jo@mein.io>
5 years agokernel: bump 4.19 to 4.19.73
Koen Vandeputte [Wed, 18 Sep 2019 10:50:22 +0000 (12:50 +0200)]
kernel: bump 4.19 to 4.19.73

Refreshed all patches.

Remove upstreamed:
- 840-media-i2c-tda1997x-select-V4L2_FWNODE.patch
- 079-v4.20-ARM-dts-qcom-ipq4019-fix-PCI-range.patch
- 083-ARM-dts-qcom-ipq4019-enlarge-PCIe-BAR-range.patch
- 084-ARM-dts-qcom-ipq4019-Fix-MSI-IRQ-type.patch

Fixes:
- CVE-2019-15030

Compile-tested on: cns3xxx
Runtime-tested on: cns3xxx

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
5 years agokernel: bump 4.14 to 4.14.144
Koen Vandeputte [Wed, 18 Sep 2019 10:49:31 +0000 (12:49 +0200)]
kernel: bump 4.14 to 4.14.144

Refreshed all patches.

Altered patches:
- 816-pcie-support-layerscape.patch

Fixes:
-CVE-2019-15030

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

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
5 years agokernel: bump 4.9 to 4.9.193
Koen Vandeputte [Wed, 18 Sep 2019 10:45:40 +0000 (12:45 +0200)]
kernel: bump 4.9 to 4.9.193

Refreshed all patches.

Fixes:
- CVE-2019-15030

Compile-tested on: none
Runtime-tested on: none

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
5 years agoramips: add support for EnGenius ESR600
Nick Briggs [Sat, 20 Jul 2019 03:01:29 +0000 (20:01 -0700)]
ramips: add support for EnGenius ESR600

The EnGenius ESR600 is a dual band wireless router with a 4-port gigabit
Ethernet switch, a gigabit Ethernet WAN port and a USB port.

Specification:

- Bootloader: U-Boot
- SoC: MediaTek MT7620A (600 MHz)
- Flash: 16MB, Macronix MX25L12845E
- RAM: 64MB, Nanya NT5TU32M16DG-AC
- Serial: 115200 baud, no header, 3.3V
   J2: Vcc (arrow), Gnd, Tx, Rx
- USB: USB 2, 5V
- Ethernet: 5 x 1 Gb/s 4 LAN 1 WAN, Atheros AR8327
- WiFi0: 5 GHz 802.11 b/g/n Ralink RT5592N
   300 Mb/s, 2T2R
- WiFi1: 2.4 GHz 802.11 b/g/n integrated
   300 Mb/s, 2T2R
- Antennas: 2 per radio, internal
- LEDs: 1 programmable power (amber)
   2 programable radio (blue)
1 programable WPS-5G (blue)
1 non-programable WAN activity (blue)
1 unconfigured WPS-2.4G (amber)
- Buttons: GPIO: Reset, WPS

Installation:

Use the OEM web interface to install the ...-factory.dlf image.
Use the OpenWRT ...-sysupgrade.bin image for future upgrades.

The J2 serial port can be accessed either by soldering in a header,
standard 0.1" spacing, or by using pogo-pins against the back side.

As configured by the OEM, the U-Boot boot delay is short, however quickly
typing "1" leads to the U-Boot "System load Linux to SDRAM via TFTP"
prompt.  The TFTP client is configured by default with
     client:   192.168.99.9
     server:   192.168.99.8
     filename: uImageESR600
It will load an OpenWRT initramfs kernel with this method.

Known issues:

1) Only the ports externally labeled WAN, LAN3 and LAN4 are operational.
LAN1 and LAN2 do not appear to power up. This issue is also present
in the Lava LR25G001.

2) The amber WPS-2.4G LED, in the same lightguide as the blue WPS-5G LED,
is not configured in the Device Tree specification.

3) The blue WAN activity LED is not configured in the Device Tree
specification as this causes the AR8327 switch to fail to initialize.

Signed-off-by: Nick Briggs <nicholas.h.briggs@gmail.com>
[merge conflict in 02_network]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agoramips: mt7621: Add new device AsiaRF AP7621-NV1
Daniel Danzberger [Mon, 22 Jul 2019 10:54:06 +0000 (12:54 +0200)]
ramips: mt7621: Add new device AsiaRF AP7621-NV1

SoC:    Mediatek MT7621A
CPU:    4x 880Mhz
Cache:  32 KB I-Cache and 32 KB D-Cach
        256 KB L2 Cache (shared by Dual-Core)
RAM:    DDR3 512MB 16bits BUS
FLASH:  16MB
Switch: Mediatek Gigabit Switch (2 x LAN, 1 x WAN)
POE:    (1x PD, 2x PSE)
USB:    1x 3.0
PCI:    3x Mini PCIe (3 USB2.0 + 2 x UIM interface)
GPS:    Quectel L70B
SIM:    2 Slots
BTN:    Reset
LED:    - Power
        - Ethernet
        - Wifi
        - USB
UART:  UART is present as Pads with throughholes on the PCB.
       They are located on left side.
       3.3V - RX - GND - TX / 57600-8N1
       3.3V is the square pad

Installation
------------
The stock image is a modified openwrt and can be overflashed via sysupgrade -F

Signed-off-by: Daniel Danzberger <daniel@dd-wrt.com>
[merge conflict in mt7621.mk]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agoramips: add support for Asus RT-AC85P
Birger Koblitz [Sun, 15 Sep 2019 14:20:11 +0000 (16:20 +0200)]
ramips: add support for Asus RT-AC85P

SoC: MediaTek MT7621AT dual-core @ 880MHz
RAM: 256M (Winbond W632GG6KB-1)
FLASH: 128MB (Macronix MX30LF1G18AC-TI)
WiFi: - 2.4GHz MediaTek MT7615N bgn
- 5GHz MediaTek MT7615N nac
Switch: SoC integrated Gigabit Switch (4 x LAN, 1 x WAN)
USB: 1 x USB 3.1 (Gen 1)
BTN: Reset, WPS
LED: - Power (blue)
- 5Ghz (blue)
- 2.4GHz (blue)
- Internet (blue)
- 4x LAN (blue)
(LAN/WAN leds are not controllable by GPIOs)
UART:  UART is present as Pads marked J4 on the PCB.
3.3V - TX - RX - GND / 57600-8N1
3.3V is the square pad
MAC: The MAC address on the router-label matches the MAC of
the 2.4 GHz WiFi.
LAN and WAN MAC are identical: MAC_LABEL+4
5 GHz WiFi MAC: also MAC_LABEL+4

Installation
------------
Via U-Boot tftpd:
Switch on device, within 2s press reset button and keep pressed
until power LED starts blinking slowly.
Upload factory image via tftp put, the router's ip is 192.168.1.1
and expects the client on 192.168.1.75.

The images also work on the Asus RT-AC65P models as tested by Gabor.

Signed-off-by: Birger Koblitz <mail@birger-koblitz.de>
Tested-by: Gabor Varga <vargagab@gmail.com>
[fixed Asus -> ASUS in DTS]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agoramips: add support for Edimax RG21S
Birger Koblitz [Thu, 19 Sep 2019 19:33:59 +0000 (21:33 +0200)]
ramips: add support for Edimax RG21S

SoC: MediaTek MT7621AT dual-core @ 880MHz
RAM: 256M (Nanya NT5CC128M)
FLASH: 16MB (Macronix MX25L12835F)
WiFi: - 2.4GHz MediaTek MT7615N bgn
- 5GHz MediaTek MT7615N nac
Switch: SoC integrated Gigabit Switch (4 x LAN, 1 x WAN)
USB: No
BTN: Reset, WPS
LED: 4 red LEDs, indistinguishable when casing closed
UART:  UART is present as Pads marked J1 on the PCB.
3.3V - RX - GND - TX / 57600-8N1
3.3V is the square pad

Installation
------------
Update the factory image via the OEM web-interface
(by default:http://192.168.1.1)
The sysupgrade image can be installed via TFTP from
the U-Boot bootloader. Connect ethernet port 2.

Signed-off-by: Birger Koblitz <mail@birger-koblitz.de>
[flash node rename, EDIMAX -> Edimax, complete device model name]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agobrcm2708: fix Hifiberry DAC+DSP package definition
Álvaro Fernández Rojas [Fri, 20 Sep 2019 05:50:47 +0000 (07:50 +0200)]
brcm2708: fix Hifiberry DAC+DSP package definition

Fixes: ce536ae ("brcm2708: add support for Hifiberry DAC+DSP")
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
5 years agokernel: Correctly search for the FIT image in mtd partition.
Fredrik Olofsson [Wed, 4 Sep 2019 08:41:22 +0000 (10:41 +0200)]
kernel: Correctly search for the FIT image in mtd partition.

Previously all iterations of the loop checked offset=0 in the partition.

Signed-off-by: Fredrik Olofsson <fredrik.olofsson@anyfinetworks.com>
5 years agobuild: make device tree arg really optional in mkits.sh
Johann Neuhauser [Mon, 16 Sep 2019 09:42:49 +0000 (11:42 +0200)]
build: make device tree arg really optional in mkits.sh

If no device tree is given there is no node generated, but
the configuration does still include the name of the missing node.
This will result in a successful build fit image, but bootm does
throw a error message if we want to boot the bad configuration.

Signed-off-by: Johann Neuhauser <jneuhauser@dh-electronics.com>
5 years agolantiq-dsl: fix value of FECS counter
Martin Schiller [Mon, 26 Aug 2019 05:37:14 +0000 (07:37 +0200)]
lantiq-dsl: fix value of FECS counter

Instead of the nFECS value, the nFEC value (the number of corrected code words)
was mistakenly used.

Signed-off-by: Martin Schiller <ms@dev.tdt.de>
5 years agokernel: add module for Emulex OneConnect 10Gbit
Alberto Bursi [Sun, 15 Sep 2019 22:54:20 +0000 (00:54 +0200)]
kernel: add module for Emulex OneConnect 10Gbit

add module to support Emulex OneConnect
common in 10Gbit SFP+ cards by Dell/HP/IBM
supports OneConnect OCe10xxx OCe11xxx OCe14xxx,
LightPulse LPe12xxx

Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
5 years agohostapd: declare struct wpa_bss early
Leon M. George [Wed, 11 Sep 2019 13:22:55 +0000 (15:22 +0200)]
hostapd: declare struct wpa_bss early

wps_supplicant.h assumes that 'struct wpa_bss' is forward declared if
CONFIG_WPS is not defined.  With the later inclusion of
600-ubus_support, the issue manifests in warnings like these:

wps_supplicant.h:113:15: warning: 'struct wpa_bss' declared inside parameter list will not be visible outside of this definition or declaration
        struct wpa_bss *bss)
               ^~~~~~~

This patch forward declares 'struct wpa_bss' regardless.

Signed-off-by: Leon M. George <leon@georgemail.eu>
[commit message facelift]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agohostapd: revert signature change in patch
Leon M. George [Wed, 11 Sep 2019 12:10:18 +0000 (14:10 +0200)]
hostapd: revert signature change in patch

The original wpa_hexdump uses a 'void *' for the payload.  With patch
410-limit_debug_messages, the signature changes and compiler warnings
occur at various places.  One such warning is:

 wpa_debug.h:106:20: note: expected 'const u8 * {aka const unsigned char *}' but argument is of type 'struct wpa_eapol_key *'

Signed-off-by: Leon M. George <leon@georgemail.eu>
[commit message facelift]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agoath79: add SUPPORTED_DEVICES to ubnt_nanostation-m-xw
Roger Pueyo Centelles [Tue, 17 Sep 2019 17:31:06 +0000 (19:31 +0200)]
ath79: add SUPPORTED_DEVICES to ubnt_nanostation-m-xw

The ar71xx images for the Ubiquiti NanoStation M (XW) devices use
"nanostation-m-xw" as the board name, but the ath79 images are only
compatible with the "nano-m-xw" board name, so sysupgrade complains.

By adding this additional supported device, sysuspgrade smoothly
upgrades from ar71xx to ath79.

Tested on a NanoStation M (XW) running OpenWrt ar71xx r10250-016d1eb.

Ref: https://github.com/openwrt/openwrt/pull/2418
Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
[removed duplicate DEVICE_VARIANT, removed uneeded nano-m-xw support]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agoath79: add XM device variant for Ubiquiti devices
Adrian Schmutzler [Wed, 18 Sep 2019 19:36:46 +0000 (21:36 +0200)]
ath79: add XM device variant for Ubiquiti devices

So far, XW devices have DEVICE_VARIANT defined and XM devices have
no DEVICE_VARIANT set.

This adds DEVICE_VARIANT for XM devices, and moves definitions for
both XM and XW to the common definitions.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agoramips: provide label MAC address
Adrian Schmutzler [Mon, 24 Jun 2019 15:57:15 +0000 (17:57 +0200)]
ramips: provide label MAC address

This patch adds the label MAC address for several devices in
ramips.

Some devices require setting the MAC address in 02_network:

For the following devices, the netif device can be linked in
device tree, but the MAC address cannot be read:
- cudy,wr1000
- dlink,dir-615-d
- dlink,dir-615-h1
- dlink,dir-860l-b1
- glinet,gl-mt300a
- glinet,gl-mt300n
- glinet,gl-mt750
- vocore,vocore2
- vocore,vocore2-lite
- zbtlink,zbt-we1326
- zbtlink,zbt-wg3526

For the following devices, label MAC address is tied to lan or
wan, so no node to link to exists in device tree:
- dlink,dir-510l
- dlink,dwr-116-a1
- dlink,dwr-118-a1
- dlink,dwr-118-a2
- dlink,dwr-921-c1
- dlink,dwr-922-e2
- all hiwifi devices
- lava,lr-25g001
- xiaomi,mir3p

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agoipq806x: provide label MAC address
Adrian Schmutzler [Thu, 15 Aug 2019 13:13:44 +0000 (15:13 +0200)]
ipq806x: provide label MAC address

This patch adds the label MAC address for several devices in
ipq806x.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agompc85xx: provide label MAC address
Adrian Schmutzler [Thu, 15 Aug 2019 13:13:38 +0000 (15:13 +0200)]
mpc85xx: provide label MAC address

This patch adds the label MAC address for some devices in
mpc85xx.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agoath79: provide label MAC address
Adrian Schmutzler [Thu, 15 Aug 2019 13:13:33 +0000 (15:13 +0200)]
ath79: provide label MAC address

This patch adds the label MAC address for several devices in
ath79.

Some devices require setting the MAC address in 02_network:

For the following devices, the netif device can be linked in
device tree, but the MAC address cannot be read:
- alfa-network,ap121f
- avm,fritz300e
- ubnt-xm devices

For the following devices, label MAC address is tied to lan or
wan, so no node to link to exists in device tree:
- adtran,bsap1800-v2
- adtran,bsap1840
- dlink,dir-842-c1/-c2/-c3
- engenius,ecb1750
- iodata,etg3-r
- iodata,wn-ac1167dgr
- iodata,wn-ac1600dgr
- iodata,wn-ac1600dgr2
- iodata,wn-ag300dgr
- nec,wg800hp
- nec,wg1200cr
- trendnet,tew-823dru

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agobase-files: provide option to specify label MAC address in board.d
Adrian Schmutzler [Thu, 15 Aug 2019 13:13:27 +0000 (15:13 +0200)]
base-files: provide option to specify label MAC address in board.d

For many devices, MAC addresses cannot be retrieved via the
device tree alias.

To still provide the label MAC address for those, this implements
a second mechanism that will put the address into uci config.
Note that this stores the actual MAC address, whereas in DTS
we reference the bearing device.

This is based on the work of Rosy Song <rosysong@rosinson.com>

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agobase-files: add function to retrieve label MAC address
Adrian Schmutzler [Thu, 15 Aug 2019 13:13:21 +0000 (15:13 +0200)]
base-files: add function to retrieve label MAC address

To refer to the MAC address on a device's label, one can
specify the alias label-mac-device in the DTS which should
point to the bearer of the corresponding MAC address.

With the function get_mac_label, the user can retrieve then
retrieve this address and use it as a value that uniquely
identifies his device.

This is severely helpful for several downstream functionalities,
e.g. define MAC addresses of custom netifs or change the SSID to
be easily recognizable.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agocurl: bump to 7.66.0
Hans Dedecker [Tue, 17 Sep 2019 20:45:41 +0000 (22:45 +0200)]
curl: bump to 7.66.0

Refresh patches, for changes in version 7.66.0 see https://curl.haxx.se/changes.html#7_66_0

Fixes CVEs:
    CVE-2019-5481
    CVE-2019-5482

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
5 years agoopenssl: bump to 1.1.1d
Eneas U de Queiroz [Tue, 17 Sep 2019 13:52:11 +0000 (10:52 -0300)]
openssl: bump to 1.1.1d

This version fixes 3 low-severity vulnerabilities:

- CVE-2019-1547: ECDSA remote timing attack
- CVE-2019-1549: Fork Protection
- CVE-2019-1563: Padding Oracle in PKCS7_dataDecode and
 CMS_decrypt_set1_pkey

Patches were refreshed.

Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
5 years agobrcm2708: add support for Hifiberry DAC+DSP
Álvaro Fernández Rojas [Thu, 19 Sep 2019 16:06:50 +0000 (18:06 +0200)]
brcm2708: add support for Hifiberry DAC+DSP

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
5 years agobrcm2708: add support for Raspberry Pi 2B 1.2
Álvaro Fernández Rojas [Thu, 19 Sep 2019 15:32:44 +0000 (17:32 +0200)]
brcm2708: add support for Raspberry Pi 2B 1.2

RPi 2B 1.2 has a BCM2710/BCM2837 like RPi 3 devices.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
5 years agobrcm2708: update to latest patches from RPi foundation
Álvaro Fernández Rojas [Thu, 19 Sep 2019 14:43:19 +0000 (16:43 +0200)]
brcm2708: update to latest patches from RPi foundation

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
5 years agobrcm2708-gpu-fw: update to latest firmware
Álvaro Fernández Rojas [Thu, 19 Sep 2019 15:37:43 +0000 (17:37 +0200)]
brcm2708-gpu-fw: update to latest firmware

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
5 years agokernel: 4.19: fix crash when setting up spi-gpio
DENG Qingfang [Thu, 19 Sep 2019 09:44:48 +0000 (11:44 +0200)]
kernel: 4.19: fix crash when setting up spi-gpio

If an spi-gpio was specified with num-chipselects = <0> in dts, kernel
will crash:

 Unable to handle kernel paging request at virtual address 32697073
 pgd = (ptrval)
 [32697073] *pgd=00000000
 Internal error: Oops: 5 [# 1] SMP ARM
 Modules linked in:
 CPU: 2 PID: 1 Comm: swapper/0 Not tainted 4.19.72 #0
 Hardware name: Generic DT based system
 PC is at validate_desc+0x28/0x80
 LR is at gpiod_direction_output+0x14/0x128
 ...
 [<c0544db4>] (validate_desc) from [<c0545228>] (gpiod_direction_output+0x14/0x128)
 [<c0545228>] (gpiod_direction_output) from [<c05fa714>] (spi_gpio_setup+0x58/0x64)
 [<c05fa714>] (spi_gpio_setup) from [<c05f7258>] (spi_setup+0x12c/0x148)
 [<c05f7258>] (spi_setup) from [<c05f7330>] (spi_add_device+0xbc/0x12c)
 [<c05f7330>] (spi_add_device) from [<c05f7f74>] (spi_register_controller+0x838/0x924)
 [<c05f7f74>] (spi_register_controller) from [<c05fa494>] (spi_bitbang_start+0x108/0x120)
 [<c05fa494>] (spi_bitbang_start) from [<c05faa34>] (spi_gpio_probe+0x314/0x338)
 [<c05faa34>] (spi_gpio_probe) from [<c05a844c>] (platform_drv_probe+0x34/0x70)

The cause is spi_gpio_setup() did not check if the spi-gpio has
chipselect pins before setting their direction and results in derefing
an invalid pointer.

The bug is spotted in kernel 4.19.72 and does not occur in 4.14.

There is a similar fix upstream in kernel 5.2 in commit 249e2632dcd0
("spi: gpio: Don't request CS GPIO in DT use-case").

Ref: https://patchwork.kernel.org/patch/11150619/
Signed-off-by: DENG Qingfang <dengqf6@mail2.sysu.edu.cn>
[use upstream patch, moved from hack to pending dir, commit facelift]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agoprocd: fix invalid JSON filter expression in procd_running()
Jo-Philipp Wich [Thu, 19 Sep 2019 05:16:49 +0000 (07:16 +0200)]
procd: fix invalid JSON filter expression in procd_running()

Since service and instance names may contain characters which are not allowed
in JSON path labels, such as dashes or spaces, change the filter expression
to array square bracket notation to properly match these cases as well.

Fixes: 2c3dd70741 ("procd: add procd_running() helper for checking running state")
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
5 years agofirewall: update to latest Git HEAD
Jo-Philipp Wich [Wed, 18 Sep 2019 08:51:24 +0000 (10:51 +0200)]
firewall: update to latest Git HEAD

383eb58 ubus: do not overwrite ipset name attribute

Ref: https://forum.openwrt.org/t/fw3-ipset-procd-objects/44044
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
5 years agoprocd: update to the latest git HEAD
Rafał Miłecki [Wed, 18 Sep 2019 05:33:01 +0000 (07:33 +0200)]
procd: update to the latest git HEAD

62dc8c0 system: sysupgrade: send reply on error
2710c65 system: refuse sysupgrade with backup if it's unsupported

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
5 years agoglibc: update to latest 2.27 commit (BZ#23637)
Hans Dedecker [Tue, 17 Sep 2019 20:11:26 +0000 (22:11 +0200)]
glibc: update to latest 2.27 commit (BZ#23637)

5b4f7382af Add undef to fix test failure.
9456483fb2 Improve performance of memmem
373f8b06a3 Improve performance of strstr
4ec1b9e913 Fix strstr bug with huge needles (bug 23637)
ecd6271ed8 Speedup first memmem match
bba6b9288f Simplify and speedup strstr/strcasestr first match
7a4da6ef7a Improve strstr performance

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
5 years agoath79: add support for TP-Link Archer C59v2
Keith Maika [Mon, 16 Sep 2019 16:45:02 +0000 (12:45 -0400)]
ath79: add support for TP-Link Archer C59v2

Add ath79 support for Archer C59v2, previously supported by ar71xx.

TP-Link Archer C59v2 is a dual-band AC1350 router based on
Qualcomm/Atheros QCA9561+QCA9886 chips.

Specification:

- 775/650/258 MHz (CPU/DDR/AHB)
- 128 MB of RAM (DDR2)
- 16 MB of FLASH (SPI NOR)
- 3T3R 2.4 GHz
- 2T2R 5 GHz
- 5x 10/100 Mbps Ethernet
- USB 2.0 port
- UART header on PCB

Flash instruction:
  - via web UI:
    1. Download openwrt-ath79-generic-tplink_archer-c59-v2-squashfs-factory.bin
    2. Login to router and open the Advanced tab
    3. Navigate to System Tools -> Firmware Upgrade
    4. Upload firmware using the Manual Upgrade form
  - via TFTP:
    1. Set PC to fixed ip address 192.168.0.66
    2. Download openwrt-ath79-generic-tplink_archer-c59-v2-squashfs-factory.bin
       and rename it to tp_recovery.bin
    3. Start a tftp server with the file tp_recovery.bin in its root directory
    4. Turn off the router
    5. Press and hold Reset button
    6. Turn on router with the reset button pressed and wait ~15 seconds
    7. Release the reset button and after a short time
       the firmware should be transferred from the tftp server
    8. Wait ~30 second to complete recovery.

Signed-off-by: Keith Maika <keithm@aoeex.com>
5 years agoath79: move common definitions from Archer C58/C59 to common DTSI
Adrian Schmutzler [Sun, 1 Sep 2019 08:59:31 +0000 (10:59 +0200)]
ath79: move common definitions from Archer C58/C59 to common DTSI

The Archer C58/C59 have redundant LED and MAC address definitions
in their DTS files. This moves them to the parent DTSI file.

The patch already accounts for the upcoming Archer C59 v2.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agobrcm47xx: sysupgrade: support Luxul firmware format
Dan Haab [Mon, 16 Sep 2019 15:18:37 +0000 (09:18 -0600)]
brcm47xx: sysupgrade: support Luxul firmware format

Allow flashing Luxul devices using vendor firmware format.

Signed-off-by: Dan Haab <dan.haab@legrand.com>
5 years agobcm53xx: sysupgrade: support Luxul firmware format
Dan Haab [Thu, 12 Sep 2019 17:14:19 +0000 (11:14 -0600)]
bcm53xx: sysupgrade: support Luxul firmware format

Allow flashing Luxul devices using vendor firmware format. The next step
will be building proper images once they are conirmed to work.

Signed-off-by: Dan Haab <dan.haab@legrand.com>
5 years agodownload.dl: fix @KERNEL mirror urls
Yousong Zhou [Mon, 16 Sep 2019 08:02:16 +0000 (08:02 +0000)]
download.dl: fix @KERNEL mirror urls

 - Prepend "/pub" for mirror.rackspace.com
 - Use https for download.xs4all.nl and mirrors.mit.edu

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
5 years agomac80211: brcmfmac: backport the last 5.4 changes
Rafał Miłecki [Mon, 16 Sep 2019 06:04:27 +0000 (08:04 +0200)]
mac80211: brcmfmac: backport the last 5.4 changes

This makes brcmfmac use the same wiphy after PCIe reset to help user
space handle corner cases (e.g. firmware crash).

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
5 years agoar71xx: fix typo in platform_do_upgrade_compex()
Rafał Miłecki [Mon, 16 Sep 2019 04:59:09 +0000 (06:59 +0200)]
ar71xx: fix typo in platform_do_upgrade_compex()

Fixes: 641f6b6c26cb ("treewide: use new procd sysupgrade $UPGRADE_BACKUP variable")
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
5 years agomediatek: enable mtk_efuse by default
Lorenzo Bianconi [Fri, 6 Sep 2019 11:14:39 +0000 (13:14 +0200)]
mediatek: enable mtk_efuse by default

Enable by default mtk_efuse driver since it needed by mtk_thermal driver
to read sensor calibration data

Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
5 years agoath79: eg200: enable status led usage
Karl Palsson [Thu, 5 Sep 2019 09:36:53 +0000 (09:36 +0000)]
ath79: eg200: enable status led usage

This enables using the "eTactica" LED during boot, to indicate failsafe,
and during upgrade, while still leaving the LED alone for normal
operation.  This brings the device more in line with how other devices
work, and makes the failsafe functionality easier to use and understand.

Signed-off-by: Karl Palsson <karlp@etactica.com>
5 years agofirewall: update to latest git HEAD
Petr Štetiar [Sun, 15 Sep 2019 19:19:00 +0000 (21:19 +0200)]
firewall: update to latest git HEAD

c26f8907d1d2 firewall3: fix typo that affects ICMPv6 rules with numeric icmp_type

Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agoath10k-ct: update to version 2019-09-09
Robert Marko [Fri, 13 Sep 2019 10:15:39 +0000 (12:15 +0200)]
ath10k-ct: update to version 2019-09-09

Update the ath10k-ct driver version to 5e8cd86f90dac966d12df6ece84ac41458d0e95f
to enable dynamic VLANs to work. Patches refreshed during the bump.

Signed-off-by: Robert Marko <robimarko@gmail.com>
[commit description facelift]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agoath10k-firmware: update Candela Tech firmware images
Robert Marko [Fri, 13 Sep 2019 10:15:37 +0000 (12:15 +0200)]
ath10k-firmware: update Candela Tech firmware images

This enables a feature flag in the wave-2 firmware wmi-services indicating it can send
software-encrypted raw frames.  This should in turn allow the AP-VLAN feature to work.

Signed-off-by: Robert Marko <robimarko@gmail.com>
5 years agoscons: move to packages feed
Petr Štetiar [Sat, 27 Jul 2019 22:33:32 +0000 (00:33 +0200)]
scons: move to packages feed

This patch removes scons host build tool, as commit 7087efd72a8d
("scons: move host build tool to a proper place") in the packages feed
has moved scons into the new home.

There are currently no packages in the master tree which would need
scons, yet scons is build always as part of host tools, just in order to
satisfy host build dependency of few packages in the packages feeds.

Ref: https://github.com/openwrt/packages/pull/9584
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agonetifd,lldpd,rpcd,log: use generic service_running
Petr Štetiar [Tue, 23 Jul 2019 10:16:03 +0000 (12:16 +0200)]
netifd,lldpd,rpcd,log: use generic service_running

commit eb204d14f75c ("base-files: implement generic service_running")
introduced generic service_running so it's not needed to copy&paste same
3 lines over and over again.

I've removed service_running from netifd/network init script as well,
because it was not working properly, looked quite strange and I didn't
understand the intention:

 $ /etc/init.d/network stop
 $ service network running && echo "yes" || echo "nope"
     ( have to wait for 30s )
 Command failed: Request timed out
 yes

Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agobase-files,procd: add generic service status
Luiz Angelo Daros de Luca [Wed, 12 Sep 2018 01:35:09 +0000 (22:35 -0300)]
base-files,procd: add generic service status

Adds a default status action for init.d scripts.

procd "service status" will return:

 0) for loaded services (even if disabled by conf or dead)
 3) for inactive services
 4) when filtering a non-existing instance

Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
[rebased, cleaned up]
Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agobase-files: implement generic service_running
Petr Štetiar [Tue, 23 Jul 2019 10:06:29 +0000 (12:06 +0200)]
base-files: implement generic service_running

DRY is good, otherwise we're going to suffer with a copy&paste disease
in the init scripts.

Signed-off-by: Petr Štetiar <ynezz@true.cz>
5 years agoodhcpd: update to latest git HEAD
Hans Dedecker [Sun, 15 Sep 2019 18:40:16 +0000 (20:40 +0200)]
odhcpd: update to latest git HEAD

1d24009 netlink: rename netlink callback handlers
91a28e4 ndp: answer global-addressed NS manually
fd93e36 dhcpv6: retry failed PD assignments on addrlist change

Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
5 years agoramips: add SD-Card support for Linkit Smart 7688
Ivan Hörler [Sat, 14 Sep 2019 15:25:04 +0000 (17:25 +0200)]
ramips: add SD-Card support for Linkit Smart 7688

The Linkit Smart 7688 has a SD-Card reader that does not work with the official build of openwrt. Adding kmod-sdhci-mt7620 makes it working.

Signed-off-by: Ivan Hörler <i.hoerler@me.com>
5 years agoramips: only add spidev node in for WrtNode2R (and not for 2P)
Daniel Golle [Sun, 15 Sep 2019 18:14:26 +0000 (20:14 +0200)]
ramips: only add spidev node in for WrtNode2R (and not for 2P)

Only the 2R version got the STM32 uC connected as 2nd SPI device.
Hence move the spidev node from mt7628an_wrtnode_wrtnode2.dtsi to
mt7628an_wrtnode_wrtnode2r.dts.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
5 years agoscripts/feeds: fix accepting "-" in feed type string
Felix Fietkau [Sun, 15 Sep 2019 17:53:17 +0000 (19:53 +0200)]
scripts/feeds: fix accepting "-" in feed type string

Fixes a syntax error in processing the type src-git-full

Signed-off-by: Felix Fietkau <nbd@nbd.name>
5 years agoapm821xx: image: remove unused kernel.dtb from IMAGES
Yousong Zhou [Tue, 10 Sep 2019 13:10:08 +0000 (13:10 +0000)]
apm821xx: image: remove unused kernel.dtb from IMAGES

It's a leftover from 2271967f ("apm821xx: utilize build ARTIFACTs")

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Acked-by: Christian Lamparter <chunkeey@gmail.com>
5 years agoimx6: install-dtb as separate IMAGE
Yousong Zhou [Thu, 12 Sep 2019 07:37:05 +0000 (07:37 +0000)]
imx6: install-dtb as separate IMAGE

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
5 years agobuild: install-dtb: fix race condition when copying dtb
Yousong Zhou [Tue, 10 Sep 2019 13:40:43 +0000 (13:40 +0000)]
build: install-dtb: fix race condition when copying dtb

Currently for at91 target, Build/install-dtb can be triggered concurrently for
multiple different TARGET_FILESYSTEMS, cp command can fail when the
target file is already open exclusively by other cp process

[ -f /builder/shared-workdir/build/build_dir/target-arm_arm926ej-s_musl_eabi/linux-at91_sam9x/at91sam9263ek-uImage -a -f /builder/shared-workdir/build/build_dir/target-arm_arm926ej-s_musl_eabi/linux-at91_sam9x/root.ubifs+fs=-m_2048_-e_126KiB_-c_2048+pkg=68b329da ]
[ -f /builder/shared-workdir/build/build_dir/target-arm_arm926ej-s_musl_eabi/linux-at91_sam9x/at91sam9263ek-uImage -a -f /builder/shared-workdir/build/build_dir/target-arm_arm926ej-s_musl_eabi/linux-at91_sam9x/root.ext4+pkg=68b329da ]
cp -fpR /builder/shared-workdir/build/build_dir/target-arm_arm926ej-s_musl_eabi/linux-at91_sam9x/linux-4.14.141/arch/arm/boot/dts/at91sam9263ek.dtb /builder/shared-workdir/build/bin/targets/at91/sam9x/openwrt-at91-sam9x-at91sam9263ek.dtb;
cp -fpR /builder/shared-workdir/build/build_dir/target-arm_arm926ej-s_musl_eabi/linux-at91_sam9x/linux-4.14.141/arch/arm/boot/dts/at91sam9263ek.dtb /builder/shared-workdir/build/bin/targets/at91/sam9x/openwrt-at91-sam9x-at91sam9263ek.dtb;
cp: cannot create regular file '/builder/shared-workdir/build/bin/targets/at91/sam9x/openwrt-at91-sam9x-at91sam9263ek.dtb': File exists
Makefile:87: recipe for target '/builder/shared-workdir/build/build_dir/target-arm_arm926ej-s_musl_eabi/linux-at91_sam9x/tmp/openwrt-at91-sam9x-at91sam9263ek-ubifs-dtb' failed
make[4]: *** [/builder/shared-workdir/build/build_dir/target-arm_arm926ej-s_musl_eabi/linux-at91_sam9x/tmp/openwrt-at91-sam9x-at91sam9263ek-ubifs-dtb] Error 1

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
5 years agoar71xx: make IRQ fixes target specific
Koen Vandeputte [Sat, 14 Sep 2019 14:21:57 +0000 (16:21 +0200)]
ar71xx: make IRQ fixes target specific

Move the IRQ fix from generic to ar71xx specific.
Other targets like ath79 have specific pathes to delete this code.
This resulted in a build failure on ath79

While at it, wipe the 4.19 version, as ar71xx will never reach this.

Fixes: 530f76708cef ("ar71xx: Fix potentially missed IRQ handling during
dispatch")

Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
5 years agobrcm63xx: remove redundant variable definition in Makefile
Adrian Schmutzler [Wed, 11 Sep 2019 18:56:48 +0000 (20:56 +0200)]
brcm63xx: remove redundant variable definition in Makefile

For devices inheriting from bcm63xx_netgear, the IMAGES variable
is overwritten with the same values as defined in the parent
definition. So, remove the unnecessary overwrite.

Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
5 years agoimages: fix boot failures on NAND with small sub pages
Jonas Gorski [Tue, 3 Sep 2019 12:16:12 +0000 (14:16 +0200)]
images: fix boot failures on NAND with small sub pages

SquashFS has a minimum block size of at least 1k, so we need to make
sure the last data block is also at least that big.

This is not an issue on NOR or SD CARD devices, since their rootfs
partitions go all the way to the end of the usable space.

But on NAND with ubiblock, the rootfs partition will be the exact space,
rounded up to LEB size. Unfortunately, some NAND chips with small sub
pages have a LEB size of x.5 kiB. This can cause the the last data block
to be less than 1k, which will cause the last block to be inaccessible,
causing boot failures as seen on MR24:

[    1.532960] block ubiblock0_3: created from ubi0:3(rootfs)
[    1.538457] ubiblock: device ubiblock0_3 (rootfs) set to be root filesystem
[    1.552847] SQUASHFS error: squashfs_read_data failed to read block 0x621472
[    1.559896] squashfs: SQUASHFS error: unable to read id index table
[    1.566474] VFS: Cannot open root device "(null)" or unknown-block(254,0): error -5

Since on most NOR devices, the start of the squashfs partition is not
aligned. Since the start of the rootfs_data partition there is dependend
on the SquashFS size, we cannot just always pad it, as the padding could
creep into the rootfs_data partition, breaking jffs2.

So fix this by ensuring a squashfs rootfs is always a multiple of 1k
only for UBI and NAND sysupgrade images.

Fixes #2460 without affecting NOR devices.

Tested-by: Russell Senior <russell@personaltelco.net>
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>