The swconfig-based b53 driver for the BCM53128 switch stopped working
after commits b2cfed48f6 (Revert "swconfig: fix Broadcom b53 support")
and e4e410733f (kernel: export switch_generic_set_link() symbol). This
rendered the 8 LAN ports of the EdgeSwitch 8XP non-functional, so the
image compilation for the device was disabled (5a1d7d8c1b).
This commit adds the kmod-dsa-b53-mdio and kmod-dsa-b53 packages
with the upstream B53 DSA driver, replacing the swconfig-based
kmod and kmod-switch-bcm53xx-mdio downstream ones that are not used by
any other device.
The 8 LAN ports of the EdgeSwitch 8XP are usable again. The 02_network
init script has been updated with the new DSA interfaces lan1 .. lan8.
Image building has been reenabled for the device, adding the usual DSA
incompatibility notice.
Tested on a Ubiquiti EdgeSwitch 8XP.
Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
Link: https://github.com/openwrt/openwrt/pull/11680
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Read WiFi calibration data via NVMEM framework. The MAC addresses are
stored inside a file on a filesystem and hence still have to be
extracted in userspace.
WiFI EEPROM extraction has already accidentally been partially removed
by commit 3e6de5d77a ("mediatek: use NVMEM framework on all Adtran
devices").
Fixes: 3e6de5d77a ("mediatek: use NVMEM framework on all Adtran devices")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
ffba75c9cd8f iptables: free xtables_match if found in need_protomatch
bf1d5fdf6234 iptables: fix regression with unintended free in need_protomatch
1aef9791a21e defaults.c: fix ipv6 flow offloading
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Increase the failsafe waiting timeout period from 2 seconds
to 4 seconds.
Since commit 29207748b in 2015 we have had a rapid LED blinking
indication for the failsafe triggering period. But the really short
timeout of 2 seconds requires snappy reaction time from the user to
notice the LED blinking and to push button inside the short window.
Relax the timeout to 4 seconds, which more easily allows to notice
the change in LED blinking and push a button.
Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
Link: https://github.com/openwrt/openwrt/pull/11852
Signed-off-by: Robert Marko <robimarko@gmail.com>
After a long time QCA has pushed an updated release of 2.9.0.1 firmware
for IPQ8074 and QCN9074, so lets update to 2.9.0.1-02146.
Sadly, still nothing new for IPQ6018.
QCA has also moved the repository where they will be posting firmware to
their CodeLinaro instance, so we move to using that and it allows us to
remove the manual download of QCN9074 board-2.bin.
Link: https://github.com/openwrt/openwrt/pull/16720
Signed-off-by: Robert Marko <robimarko@gmail.com>
MAC-addresses are stored with colons, thus they can now be referenced
using NVMEM.
Signed-off-by: David Bauer <mail@david-bauer.net>
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16310
Signed-off-by: Robert Marko <robimarko@gmail.com>
Upstream commit 9ba0cae3cac07c21c583f9ff194f74043f90d29c made FSL_IFC
visible and selectable, which means that in order for MTD_NAND_FSL_IFC
to work, it needs these two extra CONFIG options.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16717
Signed-off-by: Robert Marko <robimarko@gmail.com>
Add the Realtek RTL8922AE support to rtw89 package.
The 8922A is a 802.11be chip that can support 2/5/6GHz 160MHz.
Signed-off-by: Chukun Pan <amadeus@jmu.edu.cn>
Link: https://github.com/openwrt/openwrt/pull/16700
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
In #16396, crashes were reported on MT7620, which were introduced by enabling
CONFIG_LIST_HARDENED. The cause seems to be random memory corruption somewhere
in the kernel. Unfortunately the crash traces do not point to the real cause
of the crash in any way.
Since MT7620 is really ancient hardware that likely only has few users left,
I don't expect that anybody will invest a significant amount of time to track
down the real cause. Because of that, let's disable CONFIG_LIST_HARDENED on
this target only, and leave it enabled on all other platforms.
Fixes: #16396
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Building uboot-mediatek fails with GCC-14, uboot v2024.10 and
CONFIG_SYS_NONCACHED_MEMORY defined with error:
cmd/cache.c: In function 'do_dcache':
cmd/cache.c:57:25: error: implicit declaration of function
'noncached_set_region' [-Wimplicit-function-declaration]
This is caused by upstream commit 7d6cee2cd0e2e2507aca1e3a6fe0e2cb241a116e
("cmd: cache: Remove weak functions") as this removes weak functions in
favor of arch-specific definitions.
This patch adds the function prototype for `noncached_set_region` to
arch-specific header for ARM. It also adds an include in cmd/cache.c to
make the function available there.
Fixes: #16697
Fixes: f8c22c9bff ("uboot-mediatek: update to U-Boot 2024.10")
Signed-off-by: Jonas Jelonek <jelonek.jonas@gmail.com>
[@dangowrt refreshed patch]
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Add preinit script to rename network interfaces according to the
label provided in Device Tree.
Fixes: b6d8297510 ("mediatek: mt7622: add Adtran SmartRG SDG-841t6 device")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Two patches declared as accepted in v6.13 were already accepted for
v6.12. Fix filenames and order of patches applied.
Reported-by: Mieczyslaw Nalewaj <namiltd@yahoo.com>
Fixes: 8cc049cec2 ("generic: phy: aquantia: move accepted patches to backport-6.6")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
This driver allows retrieving MAC addresses and other factory information
via the NVMEM framework on Adtran devices.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Building HWMON as module prevents some other drivers from getting built
into the kernel, e.g. mxl-gpy. Always enable HWMON and also set
THERMAL_HWMON.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Backport newly introduced support for 'active-high' property and use
it to correctly implement polarity assignment for Aquantia PHY LEDs.
Previously the 'active-low' property was used to switch a LED PIN to
active-high ("drive VDD" in Aquantia-speak) mode.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Only add SerDes interface modes at lower speeds to host_interfaces.
Fixes: 5281033831 ("generic: net: phy: use all SerDes MAC interface modes")
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
The snapshot builds for qoriq only contain ext4 images. Add squashfs to
FEATURES to also have the buildbots build squashfs images.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Dropping the source-only flag from qoriq and adding it to the buildbots
exposed a missing kernel config symbol: CONFIG_DRM_OFDRM. Addd it as
disabled to the qoriq kernel config, as the only device currently
supported in the qoriq target does not have any display hardware.
Reported-by: Mieczyslaw Nalewaj <namiltd@yahoo.com>
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Update to the most recent stable release,
including a new job scheduler and bug fixes.
Backport a patch to assist in reworking jobserver support.
The original patch for jobserver support requires reworking
due to the function CanRunMore() and its usage
being significantly changed, including the return
becoming an integer instead of a boolean.
AcquireToken() must now be used in CanRunMore()
in order to quantify how much the job searching loop
is able to run at any time, and in order to do so
CanRunMore() cannot be a constant function anymore.
Added:
- 010-backport-gtest.patch
Manually Adjust:
- 100-make_jobserver_support.patch
Signed-off-by: Michael Pratt <mcpratt@pm.me>
Link: https://github.com/openwrt/openwrt/pull/16693
Signed-off-by: Robert Marko <robimarko@gmail.com>
Add methods to skip the building and execution of Ninja by python
in order to allow Make to execute Ninja after the configure script.
This allows the user to build Ninja only once
if they already have an older version of Ninja built.
It also allows the user to test the jobserver functionality
by having Ninja built twice (clean then compile).
Signed-off-by: Michael Pratt <mcpratt@pm.me>
Link: https://github.com/openwrt/openwrt/pull/16693
Signed-off-by: Robert Marko <robimarko@gmail.com>
Use default CXXFLAGS and LDFLAGS variables.
Directly add --verbose flag without a variable.
Split up lines for script arguments.
Delete unnecessary lines.
Signed-off-by: Michael Pratt <mcpratt@pm.me>
Link: https://github.com/openwrt/openwrt/pull/16693
Signed-off-by: Robert Marko <robimarko@gmail.com>
The blamed commit adds a upgrade recipe for nas1dual to specify the
firmware partition name, but does not actually include the recipe that
will be called.
Since it previously relied on the default one, add that one.
Fixes: d21720fa90 ("mvebu: fix default partition name")
Link: https://github.com/openwrt/openwrt/pull/16704
Signed-off-by: Robert Marko <robimarko@gmail.com>
While the qoriq target currently supports only one device, the
WatchGuard Firebox M300, the feedback from people using it is that it's
a great bang for the buck. Additionally, people suggest having no
snapshots available hurts adoption.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
Linksys devices do no pass $cmd at all.
Fixes: #16148
Fixes: 715634e6d1 ("base-files: upgrade: nand: use "cmd" argument for extracting command")
Signed-off-by: Boris Krasnovskiy <borkra@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16690
Signed-off-by: Robert Marko <robimarko@gmail.com>
The firmware partition name is specifc to ipTIME NAS1dual and should not be
set globally.
Fixes: #16148
Fixes: 6ff970bb51 ("mvebu: add support for ipTIME NAS1dual")
Signed-off-by: Boris Krasnovskiy <borkra@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16690
Signed-off-by: Robert Marko <robimarko@gmail.com>
73a9f9f857 added support. Let's use it.
Calibration data is not yet supported.
Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16256
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
