Petr Štetiar [Thu, 19 Nov 2020 15:32:46 +0000 (16:32 +0100)]
download: handle possibly invalid local tarballs
Currently it's assumed, that already downloaded tarballs are always
fine, so no checksum checking is performed and the tarball is used even
if it might be corrupted.
From now on, we're going to always check the downloaded tarballs before
considering them valid.
Klaus Kudielka [Fri, 27 Dec 2019 13:26:37 +0000 (14:26 +0100)]
mvebu: base-files: Update Turris Omnia U-Boot environment
Move the update procedure from sysupgrade to first boot, which is much
more convenient in the sysupgrade case (otherwise the environment is
always one generation behind).
Check whether we have an old U-Boot release installed, and update the
environment only if necessary.
Some notes on the U-Boot environment:
The first 9 lines are a copy of the default environment of the old U-Boot
release - only modified, to run "distro_bootcmd", in case "mmcboot" fails
to boot the factory OS.
The remaining 16 lines are a backport of the default environment of the
new U-Boot release (shipped with CZ11NIC23). The main entry point is
"distro_bootcmd", which eventually sources boot.scr. This way, we have
a unified boot protocol for all Turris Omnia revisions so far.
This commit also fixes a shortcoming of previous Turris Omnia support:
Users may install OpenWrt with the Turris Omnia in factory state
(i.e. invalid environment store). In that case, neither fw_setenv, nor
U-Boot itself, would import the default environment from the image -
screwing up the rescue system, at least!
Signed-off-by: Klaus Kudielka <klaus.kudielka@gmail.com> Reviewed-by: Tomasz Maciej Nowak <tomek_n@o2.pl> Tested-by: W. Michael Petullo <mike@flyn.org> (Turris Omnia "2020") Tested-by: Klaus Kudielka <klaus.kudielka@gmail.com> (Turris Omnia)
Klaus Kudielka [Mon, 23 Dec 2019 07:34:49 +0000 (08:34 +0100)]
mvebu: Add turris-omnia.bootscript
In contrast to the U-Boot version shipped with older versions of Turris
Omnia (CZ11NIC13, CZ11NIC20), the version shipped with Turris Omnia 2019
(CZ11NIC23) relies on the existence of /boot.scr.
Consequently, add a suitable boot script to the sysupgrade image.
Flash instructions for Turris Omnia 2019:
- Download openwrt-...-sysupgrade.img.gz, gunzip it, and copy the resulting
.img file to the root of a USB flash drive (FAT32 or ext2/3/4).
- Enter a rescue shell: Either via 5-LED reset and ssh root@192.168.1.1
on LAN port 4, or via 7-LED reset and the serial console.
- Insert the USB drive and mount it:
mkdir /mnt; mount /dev/sda1 /mnt
- Flash the OpenWrt image to eMMC:
dd if=/mnt/openwrt-...-sysupgrade.img of=/dev/mmcblk0 bs=4096 conv=fsync
- Reboot.
Flash instructions using a temporary "medkit" installation were written for
the older versions of Turris Omnia, and will *not* work on the Turris Omnia
2019.
Signed-off-by: Klaus Kudielka <klaus.kudielka@gmail.com> Reviewed-by: Tomasz Maciej Nowak <tomek_n@o2.pl> Tested-by: W. Michael Petullo <mike@flyn.org> (Turris Omnia "2020")
Paul Spooren [Thu, 16 Jul 2020 20:34:11 +0000 (10:34 -1000)]
build: use mkhash for IPK metadata checksums
When setting the option IPK_FILES_CHECKSUMS the build system stores
checksums of all package file as metadata. In combination with pkg_check
this allows to see if a package is broken, e.g. caused by bad flash.
To create those checksums the tool `sha256sum` were used while the rest
of OpenWrt uses `mkhash`, a small & fast implementation of sha256. As
the build system does not check the existence of `sha256sum` and the
stderr output is moved to /dev/null, a situation where the option is
enabled but no actual checksum are created may occur.
Instead of adding `sha256sum` as a requirement, this replaces it with
`mkhash sha256` and adapts the `sed` pipe command to fit spacing.
CC: Xu Wang <xwang1498@gmx.com> CC: Michal Hrusecky <Michal@Hrusecky.net> Signed-off-by: Paul Spooren <mail@aparcar.org>
Daniel Golle [Fri, 27 Nov 2020 01:14:31 +0000 (01:14 +0000)]
procd: update to git HEAD
3019f50 jail: leak less memory 7e01453 jail: fix segfault on missing name and refactor 5abee8f jail: fix and simplify userns uid/gid maps from OCI 4ba72ec jail: relax /etc/resolv.conf creation db5ef86 jail: don't use NULL arguments for mount syscall 19ac9df jail: don't fail if can't mount-bind /etc/resolv.conf acf36f2 jail: seteuid before clone(CLONE_NEWUSER) e40828f jail: fix typo in usage output b87984b jail: don't attempt to mount /sys with noatime b275b11 jail: enter existing cgroups namespace if given 31e0a46 jail: properly initialize timens_fd
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Daniel Golle [Fri, 27 Nov 2020 00:00:31 +0000 (01:00 +0100)]
initramfs: switch to tmpfs to fix ujail
Hauke wrote:
> We want to run some processes in the procd-ujail, this works when we
> use a SquashFS image and an overlay file system, but when we use an
> initramfs it does not work.
> [...]
> When we switch from initramfs to tmpfs, it is working, we added this
> code to target/linux/generic/other-files/init to make [it] work.
Move files to newly mounted tmpfs and then use switch_root to chroot
into new rootfs and free initramfs.
Suggested-by: Hauke Mehrtens <hauke@hauke-m.de> Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Rosen Penev [Thu, 26 Nov 2020 00:45:56 +0000 (16:45 -0800)]
libiconv-full: Makefile polishing
Added PKG_INSTALL to avoid using an explicit define Build/Compile
Added PKG_BUILD_PARALLEL for faster compilation.
Removed TARGET_CLAFGS. They are no longer necessary.
fPIC is default now. So is gnu99. -DUSE_DOS is a hack to include old
and mostly unused conversions.
Josef Schlehofer [Mon, 20 Jul 2020 13:47:31 +0000 (15:47 +0200)]
libiconv-full: update to version 1.16
- Removed following patches:
100-strip_charsets.patch - makes the full variant slim.
101-autotools.patch - this one fails to apply because it was backported
from newer versions for 1.11.1.
103-configure_ac_fix.patch - backported from newer versions
200-work-with-libtool2.patch - is not needed anymore, it is done
differently in upstream
300-fortify-source-compat.patch - these files are not there anymore
- TVHeadend requires working iconv library e.g. transliteration to ASCII
and this does not work with libiconv-full currently.
There is a simple test, which requires to install iconv package.
Before applying this update:
root@turris:/# echo ŽluťoučkýKůň | iconv -t ASCII//TRANSLIT//IGNORE
luoukK
After applying this update:
root@turris:~# echo ŽluťoučkýKůň | iconv -t ASCII//TRANSLIT//IGNORE
Zlutouck'yKun
- Makefile changes:
Use HTTPS for their website
Fixed deprecated SPDX License Identifier
Move PKG_MAINTAINER above PKG_LICENSE
Signed-off-by: Josef Schlehofer <pepe.schlehofer@gmail.com> Tested-by: Rosen Penev <rosenp@gmail.com> [malta]
Rui Salvaterra [Wed, 25 Nov 2020 21:28:17 +0000 (21:28 +0000)]
tools/sstrip: update to latest version
Drop our local sstrip copy and use the current ELFKickers upstream
version.
Patch the original makefile in order to avoid building elftoc, since it
fails with musl's elf.h. This is fine, since we only need sstrip anyway.
Finally, add the possibility to pass additional arguments to sstrip and
pass -z (remove trailing zeros) by default, which matches the behaviour
of the previous version.
Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
[shorten long commit msg lines] Signed-off-by: Paul Spooren <mail@aparcar.org>
Since we are using mac80211 5.8, let's also switch the ath10k-ct driver
to the new 5.8 version.
Modify patches so they patch the new ath10k-ct driver version.
Adapt 164-ath10k-commit-rates-from-mac80211.patch.
Drop upstreamed 205-ath10k-Add-NL80211_EXT_FEATURE_AQL-flag.patch.
Drop the other options for CT_KVER from the comment, as it is incorrect
and there are too many versions to sum up and maintain there.
Runtime-tested on ath79 (D-Link DAP-2695-A1, TP-Link EAP245-v3).
Strictly, an SPDX identifier requires a space between the comment
marker and the identifier itself. The choice of the comment marker
itself is irrelevant.
Correct:
// SPDX-License-Identifier: GPL-2.0-or-later OR MIT
Wrong:
//SPDX-License-Identifier: GPL-2.0-or-later OR MIT
Fix that in the whole tree (actually, only ramips contained wrong
uses).
Found by checkpatch.pl
Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Ataberk Özen [Wed, 19 Aug 2020 18:43:54 +0000 (21:43 +0300)]
ramips: add support for Xiaomi Mi Router 4C
This commit adds support for Xiaomi's Mi Router 4C device.
Specifications:
- CPU: MediaTek MT7628AN (580MHz)
- Flash: 16MB
- RAM: 64MB DDR2
- 2.4 GHz: IEEE 802.11b/g/n with Integrated LNA and PA
- Antennas: 4x external single band antennas
- WAN: 1x 10/100M
- LAN: 2x 10/100M
- LEDs: 2x yellow/blue. Programmable (labelled as power on case)
- Non-programmable (shows WAN activity)
- Button: Reset
How to install:
1- Use OpenWRTInvasion to gain telnet and ftp access.
2- Push openwrt firmware to /tmp/ using ftp.
3- Connect to router using telnet. (IP: 192.168.31.1 -
Username: root - No password)
4- Use command "mtd -r write /tmp/firmware.bin OS1" to flash into
the router..
5- It takes around 2 minutes. After that router will restart itself
to OpenWrt.
Signed-off-by: Ataberk Özen <ataberkozen123@gmail.com>
[wrap commit message, bump PKG_RELEASE for uboot-envtools, remove
dts-v1 from DTS, fix LED labels] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
[add "redundant" to title, remove declaration of magic variable] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Rosen Penev [Thu, 15 Oct 2020 05:58:26 +0000 (22:58 -0700)]
base-files: functions.sh: replace [^...] with [!...] in case
Strictly speaking, ash does not support it.
From https://wiki.ubuntu.com/DashAsBinSh#A.5B.5E.5D
Not to be confused by sed's and other program's regular expression
syntax. Uses of [^...] in case (parameter/word expansion in general) need
to be replaced with [!...].
Found with shellcheck: https://github.com/koalaman/shellcheck/wiki/SC2169
Sander Vanheule [Wed, 4 Nov 2020 09:21:16 +0000 (10:21 +0100)]
ath79: netgear: trim down uImage customisations
Replace NETGEAR_KERNEL_MAGIC by UIMAGE_MAGIC to better match the
variable's purpose. This allows to drop the custom
Build/netgear-uImage.
Signed-off-by: Sander Vanheule <sander@svanheule.net>
[keep UIMAGE_MAGIC definitions even for default value] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Sander Vanheule [Wed, 4 Nov 2020 09:21:15 +0000 (10:21 +0100)]
build: add UIMAGE_MAGIC to device variables
Allow a device recipe to specify a custom UIMAGE_MAGIC value, as used by
OpenWrt's -M flag for mkimage. This allows to automatically customize
the magic bytes in all calls to Build/uImage for this device, similar to
the behaviour of UIMAGE_NAME. Since the -M argument is inserted before
the user arguments, it can be overriden.
The following example would use 0x87654321 for the KERNEL image, but
0x12345678 for the KERNEL_INITRAMFS image:
Sander Vanheule [Wed, 4 Nov 2020 09:21:14 +0000 (10:21 +0100)]
ramips: mt7621: drop custom uImage function
Use the mkimage argument overrides provided by uImage to implement the
customisations required for the initramfs, instead of the near-identical
custom function.
Sander Vanheule [Wed, 4 Nov 2020 09:21:13 +0000 (10:21 +0100)]
build: uImage: allow override of default arguments
For some build recipes, the argument to Build/uImage is used to sneak in
extra arguments for mkimage, whereas this appears to have been intended
to specificy the compression method only.
Use the first provided word for -C to be backwards compatible with
current calls to Build/uImage. Use the rest of the call arguments to
override the provided defaults. Only the input file name (-d) and the
output file name cannot overriden.
Sort patches according to target/linux/generic/PATCHES. Additionally:
- replace hashes in backported patches with the ones from main Linux tree
- add descriptions to some patches
Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
[remove 004-add_sata_disk_activity_trigger.patch separately] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Filip Moc [Tue, 24 Nov 2020 21:15:26 +0000 (22:15 +0100)]
ramips: add support for TP-Link MR6400 v5
TL-MR6400v5 is very similar to TL-MR6400v4. Main differences are:
- smaller form factor
- different LED GPIOs
- different switch connections
You can flash via tftp recovery:
- serve tftp-recovery image as /tp_recovery.bin on 192.168.0.225/24
- connect to any ethernet port
- power on the device while holding the reset button
- wait at least 8 seconds before releasing reset button
Flashing via OEM web interface does not work.
LTE module does not support DHCP so it must be configured via QMI.
phy0 *:d3 art 0x1002 (label)
eth0 *:d4 art 0x0/0x6
**Installation:**
- if you get Failsafe Mode from failed flash:
only use it to flash Original firmware from Engenius
or risk kernel loop or halt which requires serial cable
Method 1: Firmware upgrade page:
OEM webpage at 192.168.1.1
username and password "admin"
Navigate to "Firmware" page from left pane
Click Browse and select the factory.bin image
Upload and verify checksum
Click Continue to confirm and wait 3 minutes
Method 2: Serial to load Failsafe webpage:
After connecting to serial console and rebooting...
Interrupt uboot with any key pressed rapidly
execute `run failsafe_boot` OR `bootm 0x9fdf0000`
wait a minute
connect to ethernet and navigate to
"192.168.1.1/index.htm"
Select the factory.bin image and upload
wait about 3 minutes
**Return to OEM:**
If you have a serial cable, see Serial Failsafe instructions
*DISCLAIMER*
The Failsafe image is unique to Engenius boards.
If the failsafe image is missing or damaged this will not work
DO NOT downgrade to ar71xx this way, can cause kernel loop or halt
The easiest way to return to the OEM software is the Failsafe image
If you dont have a serial cable, you can ssh into openwrt and run
`mtd -r erase fakeroot`
Wait 3 minutes
connect to ethernet and navigate to 192.168.1.1/index.htm
select OEM firmware image from Engenius and click upgrade
**TFTP recovery** (unstable / not reliable):
rename initramfs to 'vmlinux-art-ramdisk'
make available on TFTP server at 192.168.1.101
power board while holding or pressing reset button repeatedly
NOTE: for some Engenius boards TFTP is not reliable
try setting MTU to 600 and try many times
**Format of OEM firmware image:**
The OEM software of EAP300 v2 is a heavily modified version
of Openwrt Kamikaze. One of the many modifications
is to the sysupgrade program. Image verification is performed
simply by the successful ungzip and untar of the supplied file
and name check and header verification of the resulting contents.
To form a factory.bin that is accepted by OEM Openwrt build,
the kernel and rootfs must have specific names
and begin with the respective headers (uImage, squashfs).
Then the files must be tarballed and gzipped.
The resulting binary is actually a tar.gz file in disguise.
This can be verified by using binwalk on the OEM firmware images,
ungzipping then untaring.
The OEM upgrade script is at /etc/fwupgrade.sh.
OKLI kernel loader is required because the OEM software
expects the kernel size to be no greater than 1536k
and otherwise the factory.bin upgrade procedure would
overwrite part of the kernel when writing rootfs.
Signed-off-by: Michael Pratt <mcpratt@pm.me>
[clarify MAC address section, bump PKG_RELEASE for uboot-envtools] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Rui Salvaterra [Wed, 25 Nov 2020 09:38:51 +0000 (09:38 +0000)]
apm821xx: clean up target/subtarget features
Move features common to all subtargets to the parent target.
Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
[split patch to make it target-specific, adjust commit title/message] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Karel Kočí [Mon, 19 Oct 2020 10:32:49 +0000 (12:32 +0200)]
ath10k-ct-firmware: add conflicts and better provides
This expands packages to define not only provides but also conflicts.
These packages provides same files so they should specify conflicts.
Second expansion is that *-ct-htt and *-ct-full-htt firmwares can also
provide *-ct variant as that allows explicit dependency on CT variant
with various firmware modifications.
Signed-off-by: Karel Kočí <karel.koci@nic.cz>
[Bump PKG_RELEASE and format PROVIDES/CONFLICTS] Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Paul Spooren [Wed, 25 Nov 2020 08:06:37 +0000 (22:06 -1000)]
opkg: purge package from cache on hash mismatch
61b3c62 opkg_verify_integrity: better logging and error conditions f73d42f download: purge cached packages that have incorrect checksum 1c1480e download: factor out the logic for building cache filenames 293b1ce libopkg: factor out checksum and size verification a786e25 download: remove compatibility with old cache naming scheme
Paul Spooren [Mon, 16 Nov 2020 19:26:02 +0000 (09:26 -1000)]
scripts: add size_compare.sh
As package size changes are a continuous topic on the mailing list this
scripts helps developers to compare their local package modifications
against latest upstream.
The script downloads the latest package indexes based on env variables
or the `.config` file. The script compares the actual installed size
(data.tar.gz) or the IPK package size.
Checking configuration difference
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 554 100 554 0 0 336 0 0:00:01 0:00:01 --:--:-- 336
Fetching latest package indexes...
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 80634 100 80634 0 0 33499 0 0:00:02 0:00:02 --:--:-- 33485
% Total % Received % Xferd Average Speed Time Time Time Current
Dload Upload Total Spent Left Speed
100 54082 100 54082 0 0 24252 0 0:00:02 0:00:02 --:--:-- 24252
Matthew Gyurgyik [Fri, 13 Nov 2020 19:21:29 +0000 (11:21 -0800)]
imagebuilder: fix partition signature
When building images with the imagebuilder, the partition signature
never changes. The signature is generated by hashing SOURCE_DATE_EPOCH
and LINUX_VERMAGIC which are undefined. Prepopulate these variables, as
done by the SDK.
Signed-off-by: Matthew Gyurgyik <matthew@gyurgyik.io>
Martin Schiller [Wed, 28 Aug 2019 05:24:33 +0000 (07:24 +0200)]
uqmi: set plmn only if necessary
Setting the plmn to '0' (auto) will implicitly lead to a (delayed)
network re-registration, which could further lead to some timing
related issues in the qmi proto handler.
On the other hand, if you switch back from manual plmn selection
to auto mode you have to set it to '0', because this setting is
permanently "saved" in the wwan module.
Conclusion:
If plmn is configured, check if it's already set euqally in the module.
If so, do nothing. Otherwise set it.
Signed-off-by: Martin Schiller <ms@dev.tdt.de> Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Removed since could be reverse-applied by quilt and found to be included upstream:
ipq806x/patches-5.4/096-PCI-qcom-Make-sure-PCIe-is-reset-before-init-for-rev.patch
Signed-off-by: John Audia <graysky@archlinux.us>
[refresh altered targets after rebase] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Flashing instructions:
* ssh into target device and run `cliclientd stopcs`
* Upgrade with factory image via web interface
Debricking:
* Serial port can be soldered on PCB J3 (1: TXD, 2: RXD, 3: GND, 4: VCC)
* Bridge unpopulated resistors R225 (TXD) and R237 (RXD).
Do NOT bridge R230.
* Use 3.3V, 115200 baud, 8n1
* Interrupt bootloader by holding CTRL+B during boot
* tftp initramfs to flash via LuCI web interface
setenv ipaddr 192.168.1.1 # default, change as required
setenv serverip 192.168.1.10 # default, change as required
tftp 0x80800000 initramfs.bin
bootelf $fileaddr
MAC addresses:
MAC address (as on device label) is stored in device info partition at
an offset of 8 bytes. ath9k device has same address as ethernet, ath10k
uses address incremented by 1.
From OEM boot log:
Using interface ath0 with hwaddr b0:...:3e and ssid "..."
Using interface ath10 with hwaddr b0:...:3f and ssid "..."
TP-Link EAP225-Outdoor v1 is an AC1200 (802.11ac Wave-2) pole or wall
mount access point. Debricking requires access to the serial port, which
is non-trivial.
Flashing instructions:
* ssh into target device with recent (>= v1.6.0) firmware
* run `cliclientd stopcs` on target device
* upload factory image via web interface
Debricking:
To recover the device, you need access to the serial port. This requires
fine soldering to test points, or the use of probe pins.
* Open the case and solder wires to the test points: RXD, TXD and TPGND4
* Use a 3.3V UART, 115200 baud, 8n1
* Interrupt bootloader by holding ctrl+B during boot
* upload initramfs via built-in tftp client and perform sysupgrade
setenv ipaddr 192.168.1.1 # default, change as required
setenv serverip 192.168.1.10 # default, change as required
tftp 0x80800000 initramfs.bin
bootelf $fileaddr
MAC addresses:
MAC address (as on device label) is stored in device info partition at
an offset of 8 bytes. ath9k device has same address as ethernet, ath10k
uses address incremented by 1.
From stock ifconfig:
ath0 Link encap:Ethernet HWaddr D8:...:2E
ath10 Link encap:Ethernet HWaddr D8:...:2F
br0 Link encap:Ethernet HWaddr D8:...:2E
eth0 Link encap:Ethernet HWaddr D8:...:2E
Tested by forum user PolynomialDivision on firmware v1.7.0.
UART access tested by forum user arinc9.
Flashing instructions:
* Upgrade the device to firmware v1.4.0 if necessary
* Exploit the user management page in the web interface to start telnetd
by changing the username to `;/usr/sbin/telnetd -l/bin/sh&`.
* Immediately change the malformed username back to something valid
(e.g. 'admin') to make ssh work again.
* Use the root shell via telnet to make /tmp world writeable (chmod 777)
* Extract /usr/bin/uclited from the device via ssh and apply the binary
patch listed below. The patch is required to prevent `uclited -u` in
the last step from crashing.
* Copy the patched uclited programme back to the device at /tmp/uclited
(via ssh)
* Upload the factory image to /tmp/upgrade.bin (via ssh)
* Run `chmod +x /tmp/uclited && /tmp/uclited -u` to install OpenWrt.
Debricking:
* Serial port can be soldered on PCB J3 (1: TXD, 2: RXD, 3: GND, 4: VCC)
* Bridge unpopulated resistors R225 (TXD) and R237 (RXD).
Do NOT bridge R230.
* Use 3.3V, 115200 baud, 8n1
* Interrupt bootloader by holding CTRL+B during boot
* tftp initramfs to flash via the LuCI web interface
setenv ipaddr 192.168.1.1 # default, change as required
setenv serverip 192.168.1.10 # default, change as required
tftp 0x80800000 initramfs.bin
bootelf $fileaddr
Tested on the EAP245 v1 running the latest firmware (v1.4.0). The binary
patch might not apply to uclited from other firmware versions.
EAP245 v1 device support was originally developed and maintained by
Julien Dusser out-of-tree. This patch and "ath79: prepare for 1-port
TP-Link EAP2x5 devices" are based on that work.
TP-Link has developed a number of access points based on the AP152
reference board. In the EAP-series of 802.11ac access points, this
includes the following devices with one ethernet port:
* EAP225 v1/v2
* EAP225 v3
* EAP225-Outdoor v1
* EAP245 v1
Since the only differences between these devices are the ath10k wireless
radios and LEDs, a common base is provided for the overlapping support
requirements.
Hardware commonalities:
* SoC: QCA9563-AL3A MIPS 74kc v5.0 @ 775MHz, AHB @ 258MHz
* RAM: 128MiB DDR2 @ 650MHz
* Flash: 16MiB SPI NOR
* Wi-Fi 2.4GHz: provided by SoC
* Wi-Fi 5Ghz: ath10k chip on PCIe
* Ethernet: AR8033-AL1A, one 1GbE port (802.3at PoE)
This commit add a workaround for non working SGMII link observed on some
QCA956x SoCs. The workaround originates part from the U-Boot source code
from QCA, part from the implementation from TP-Link found in the GPL
tarball for the EAP245v1.
Auto calibration is done on u-boot, but always fall back to default value
0x7. Add a DTS entry serdes-cal in case a device require another value.
Signed-off-by: Julien Dusser <julien.dusser@free.fr>
[Sander Vanheule:
Minor code style fixes,
Remove hunk adding qca956x-serdes-fixup to a missing DTS,
Remove variable err that was only assigned,
Rename function to sgmii_serdes_init,
Lower priority of serdes call message to pr_debug] Signed-off-by: Sander Vanheule <sander@svanheule.net>
Some bootloaders do not set up gmac0 properly, leaving it disconnected
from the sgmii interface. If the user specificies phy-mode sgmii, then
use the gmac-config/device node to ensure the mux is configured
correctly.
Vladimir Vid [Mon, 1 Jun 2020 15:51:37 +0000 (17:51 +0200)]
mvebu: add initial support for Globalscale ESPRESSObin-Ultra
This patch adds support for Globalscale ESPRESSObin-Ultra. Device uses
the same Armada-3720 SoC with extended hardware support.
- SoC: Armada-3720
- RAM: 1 GB DDR4
- Flash: 4MB SPI NOR (mx25u3235f) + 8 GB eMMC
- Ethernet: Topaz 6341 88e6341 (4x GB LAN + 1x WAN with 30W PoE)
- WiFI: 2x2 802.11ac Wi-Fi marvell (88w8997 PCIe+USB)
- 1x USB 2.0 port
- 1x USB 3.0 port
- 1x microSD slot
- 1x mini-PCIe slot (USB [with nano-sim slot])
- 1x mini-USB debug UART
- 1x RTC Clock and battery
- 1x reset button
- 1x power button
- 4x LED (RGBY)
- Optional 1x M.2 2280 slot
** Installation **
Copy dtb from build_dir to bin/ and run tftpserver there:
$ cp ./build_dir/target-aarch64_cortex-a53_musl/linux-mvebu_cortexa53/
linux-5.4.65/arch/arm64/boot/dts/marvell/armada-3720-espressobin-ultra.dtb
bin/targets/mvebu/cortexa53/
$ in.tftpd -L -s bin/targets/mvebu/cortexa53/
Connect to the device UART via microUSB port on the back side and power on the device.
Power on the device and hit any key to stop the autoboot.
Set serverip (host IP) and ipaddr (any free IP address on the same subnet), e.g:
$ setenv serverip 192.168.1.10 # Host
$ setenv ipaddr 192.168.1.15 # Device
Ping server to confirm network is working:
$ ping $serverip
Using neta@30000 device
host 192.168.1.15 is alive
Sven Eckelmann [Sun, 22 Nov 2020 00:17:35 +0000 (01:17 +0100)]
kernel: mtd: parser: cmdline: Fix parsing of part-names with colons
Some devices (especially QCA ones) are already using hardcoded partition
names with colons in it. The OpenMesh A62 for example provides following
mtd relevant information via cmdline:
Such a partition list cannot be parsed and thus the device fails to boot.
Avoid this behavior by making sure that the start of the first part-name
("(") will also be the last byte the mtd-id split algorithm is using for
its colon search.
Fixes: d6a9a92e3217 ("kernel: bump 5.4 to 5.4.69") Signed-off-by: Sven Eckelmann <sven@narfation.org>
Sven Eckelmann [Sun, 22 Nov 2020 19:09:27 +0000 (20:09 +0100)]
ipq40xx: DTS style updates for OpenMesh devices
The OpenMesh related files were not updated since a while and the new
coding style requirements weren't integrated. This can cause problems
for new devices when an author uses these files as starting point.
* use SPDX-License-Identifiers instead of full license texts
* drop linux,default-trigger with value default-off for LEDs
* led nodes with label "abc:xyz" should have name "xyz_abc"
* led DT labels for "xyz_abc" should be "led_xyz_abc"
* "m25p80@0" flash node should be renamed to "flash@0"
* drop unnecessary empty lines
Signed-off-by: Sven Eckelmann <sven@narfation.org>
[minor commit title and message adjustments] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
Felix Fietkau [Mon, 23 Nov 2020 11:51:58 +0000 (12:51 +0100)]
netifd: update to the latest version
213748a9bcd9 system-linux: implement full device present state management for force-external devices 3abe1fc87151 system-linux: add retry for adding member devices to a bridge
Filip Moc [Wed, 18 Nov 2020 19:38:21 +0000 (20:38 +0100)]
ramips: add support for TP-Link MR6400 v4
You can flash via tftp recovery:
- serve tftp-recovery image as /tp_recovery.bin on 192.168.0.225/24
- connect to any ethernet port
- power on the device while holding the reset button
- wait at least 8 seconds before releasing reset button
Flashing via OEM web interface does not work.
LTE module does not support DHCP so it must be configured via QMI.
Filip Moc [Wed, 18 Nov 2020 19:36:12 +0000 (20:36 +0100)]
uqmi: add support for IPv4 autoconf from QMI
There already was an option for autoconfiguring IPv4 from QMI but this
was removed by commit 3b9b963e6e08 ("uqmi: always use DHCP for IPv4").
DHCP does not work on MR400 LTE module (in TL-MR6400 v4) so let's readd
support for IPv4 autoconf from QMI but this time allow to configure this
for IPv4 and IPv6 independently and keep DHCP default on IPv4.
Thomas Richard [Thu, 8 Oct 2020 12:35:24 +0000 (14:35 +0200)]
uqmi: wait forever registration if timeout set to 0
Give possibility to wait forever the registration by setting timeout
option to 0.
No timeout can be useful if the interface starts whereas no network is
available, because at the end of timeout the interface will be stopped
and never restarted.
Signed-off-by: Thomas Richard <thomas.richard@kontron.com>
Daniel Golle [Sun, 22 Nov 2020 15:20:35 +0000 (15:20 +0000)]
Revert "refpolicy: add variant that builds modular policy"
This reverts commit 9eb9943f82e0b2d5e32ffe1c63f5a82caca5094d.
Building the 'modular' variant requires 'semodule_package' from
'selinux-python' to be installed on the buildhost.
Apart from that, this change also broke the monolithic refpolicy
'targeted' build.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Felix Fietkau [Sat, 21 Nov 2020 09:23:54 +0000 (10:23 +0100)]
netifd: update to the latest version
351d690f1a09 wireless: fix passing bridge name for vlan hotplug pass-through c1c2728946b5 config: initialize bridge and bridge vlans before other devices 5e18d5b9ccb1 interface: do not force link-ext hotplug interfaces to present by default 4544f026bb09 bridge-vlan: add support for defining aliases for vlan ids
The wcsnrtombs function in all musl libc versions up through 1.2.1 has
been found to have multiple bugs in handling of destination buffer
size when limiting the input character count, which can lead to
infinite loop with no forward progress (no overflow) or writing past
the end of the destination buffera.
This function is not used internally in musl and is not widely used,
but does appear in some applications. The non-input-limiting form
wcsrtombs is not affected.
All users of musl 1.2.1 and prior versions should apply the attached
patch, which replaces the overly complex and erroneous implementation.
The upcoming 1.2.2 release will adopt this new implementation.
David Bauer [Wed, 18 Nov 2020 15:02:23 +0000 (16:02 +0100)]
scripts: download.pl: retry download using filename
With this commit, the download script will try downloading source files
using the filename instead of the url-filename in case the previous
download attempt using the url-filename failed.
This is required, as the OpenWrt sources mirrors serve files using the
filename files might be renamed to after downloading. If the original
mirror for a file where url-filename and filename do not match goes
down, the download failed prior to this patch.
Further improvement can be done by performing this only for the
OpenWrt sources mirrors.
projects / thirdparty / openwrt.git / log
