Pavel Kubelun [Wed, 16 Apr 2025 18:07:32 +0000 (21:07 +0300)]
kernel: r8125: disable ASPM
Disable ASPM support for this NIC, fixing strange behavior problems, such as
increased latency, strange uneven throughput, etc.
With this option disabled the NIC achieves stable performance.
Upsteam r8169 driver disables ASPM by default for this NIC.
This allows the network interface naming to be stable, free from any
possible interaction from external USB network devices that might
claim usb* interface names.
This allows the network interface naming to be stable, free from any
possible interaction from external USB network devices that might
claim usb* interface names.
This allows the network interface naming to be stable, free from any
possible interaction from external USB network devices that might
claim usb* interface names.
This allows the network interface naming to be stable, free from any
possible interaction from external USB network devices that might
claim usb* interface names.
This allows the network interface naming to be stable, free from any
possible interaction from external USB network devices that might
claim usb* interface names.
John Audia [Wed, 12 Feb 2025 13:56:47 +0000 (08:56 -0500)]
openssl: update to 3.0.16
Changes between 3.0.15 and 3.0.16 [11 Feb 2025]
CVE-2024-13176[1] - Fixed timing side-channel in ECDSA signature
computation.
There is a timing signal of around 300 nanoseconds when the top word of
the inverted ECDSA nonce value is zero. This can happen with significant
probability only for some of the supported elliptic curves. In
particular the NIST P-521 curve is affected. To be able to measure this
leak, the attacker process must either be located in the same physical
computer or must have a very fast network connection with low latency.
CVE-2024-9143[2] - Fixed possible OOB memory access with invalid
low-level GF(2^m) elliptic curve parameters.
Use of the low-level GF(2^m) elliptic curve APIs with untrusted explicit
values for the field polynomial can lead to out-of-bounds memory reads
or writes. Applications working with "exotic" explicit binary (GF(2^m))
curve parameters, that make it possible to represent invalid field
polynomials with a zero constant term, via the above or similar APIs,
may terminate abruptly as a result of reading or writing outside of
array bounds. Remote code execution cannot easily be ruled out.
Robert Marko [Fri, 2 May 2025 09:07:54 +0000 (11:07 +0200)]
toolchain: binutils: fix compilation with GCC15
GCC15 has switched the C language default from GNU17 to GNU23[1] and this
causes builds to fail with:
In file included from mips-opc.c:29:
mips-opc.c: In function 'decode_mips_operand':
mips-formats.h:86:7: error: expected identifier or '(' before 'static_assert'
86 | static_assert[(1 << (SIZE)) == ARRAY_SIZE (MAP)]; \
| ^~~~~~~~~~~~~
mips-opc.c:214:15: note: in expansion of macro 'MAPPED_REG'
214 | case 'z': MAPPED_REG (0, 0, GP, reg_0_map);
| ^~~~~~~~~~
So, backport upstream fix for this[2] to fix compilation with GCC15.
Patch for 2.40 was manually refreshed as part of the S390 code does not
exist in 2.40 as it was added after it.
Robert Marko [Wed, 16 Apr 2025 12:04:26 +0000 (14:04 +0200)]
tools: gmp: fix compilation with GCC15
Fedora 42 updated to GCC15 which now defaults to GNU23 as the default
instead of GNU17[1], and this breaks GMP compilation by failing to find
a working compiler test.
Its been fixed upstream [2][3], so backport the fix to fix GCC15 compilation.
Nick Hainke [Sun, 4 May 2025 20:38:10 +0000 (22:38 +0200)]
tools/b43-tools: update to latest version, fix C23 build
Update b43-tools to the latest upstream version to fix a
compilation error with C23:
util.h:25:15: error: 'bool' cannot be defined via 'typedef'
25 | typedef _Bool bool;
| ^~~~
Changelog: c6fc53f replace custom bool typedef with <stdbool.h> dadf30c fix format warning in compilation 2fe10ea b43-fwdump: Fix forwarding of arguments to disassembler
Hauke Mehrtens [Sun, 18 May 2025 13:09:41 +0000 (15:09 +0200)]
tools: elfutils: fix compilation with GCC15
Backport fix from upstream to fix compilation with host GCC 15.
https://sourceware.org/git/?p=elfutils.git;a=commitdiff_plain;h=7508696d107ca01b65ce8273c881462a8658f90f
Robert Marko [Sat, 24 May 2025 10:20:54 +0000 (12:20 +0200)]
image: only filter out images when ImageBuilder is used
Currently, we are filtering out images if DEFAULT:=n or BROKEN:=y are set,
so if you are building from scratch and want to build custom images that
are stripped down to fit, you must edit the image recipe or its just
filtered out.
So, to allow this behaviour when building from scratch as we can assume
that person doing that knows what they are attempting to do lets just limit
the filtering to ImageBuilder.
Fixes: f060615a78e5 ("image: respect DEFAULT and BROKEN when Default profile is selected") Signed-off-by: Robert Marko <robimarko@gmail.com>
(cherry picked from commit d0d1f190683ad5923182140c40afb479907a1c2b)
Robert Marko [Fri, 16 May 2025 11:18:46 +0000 (13:18 +0200)]
image: respect DEFAULT and BROKEN when Default profile is selected
Currently, when you select the Default profile it does not honor DEFAULT:=n
nor BROKEN:=y in device profiles but rather just tries to build all of them.
This may work when building directly, but when using Image Builder it will
always fail since no kernel or anything else is present for devices that
have DEFAULT:=n or BROKEN:=Y set since those are skipped during build.
So, lets look for DEFAULT being set to "n" or BROKEN being set to "y" and
then remove clear _PROFILE_SET so they dont end up being marked for
installation.
Eric Fahlgren [Thu, 8 May 2025 22:28:52 +0000 (15:28 -0700)]
imagebuilder: exclude metadata for profiles that have no kernel
Device profiles that specify 'DEFAULT := n' are being included
in the imagebuilder metadata, specifically in .profiles.mk, even
though there is no kernel built for the device. This results in
'make info' showing the device as valid, but then 'make image
PROFILE=xxx' failing with 'No rule to make target xxx-kernel.bin ...'
We exclude these profiles from the imagebuilder, avoiding these
errors.
Antony Kolitsos [Sat, 17 May 2025 18:18:55 +0000 (11:18 -0700)]
kernel: Update to version 5.15.182
Manually adapted the following patches:
generic/backport-5.15/828-v6.4-0002-of-Update-of_device_get_modalias.patch
generic/backport-5.15/828-v6.4-0004-of-Move-of_modalias-to-module.c.patch
* Update Mozilla certificate authority bundle to version 2.70.
The following certificate authorities were added (+):
+ Telekom Security TLS ECC Root 2020
+ Telekom Security TLS RSA Root 2023
+ FIRMAPROFESIONAL CA ROOT-A WEB
+ TWCA CYBER Root CA
+ SecureSign Root CA12
+ SecureSign Root CA14
+ SecureSign Root CA15
The following certificate authorities were removed (-):
- Security Communication Root CA (closes: #1063093)
Antony Kolitsos [Thu, 8 May 2025 23:20:58 +0000 (16:20 -0700)]
generic: drop extra-old-deprecated pending fix patch for sch codel
Remove patch 620-net_sched-codel-do-not-defer-queue-length-update from openwrt-23.05 as well.
Patch is actually an ancient patch that somehow manage to be ported for 7 solid years.
This comes from [1] where a fix patch was proposed. Nobody notice that the proposed patch was actually rejected upstream in favor of [2]. And the upstream fix patch is present in kernel from version 4.18.
This means that we were actually fixing for a non existant bug and maybe introducing regression down the line.
Drop the patch for good as we already have a fix for it in flace for a long time.
Rudy Andram [Wed, 12 Feb 2025 19:28:49 +0000 (19:28 +0000)]
wireless-regdb: Update to version 2025.02.20
b43aeb5 wireless-regdb: assert and correct maximum bandwidth within frequency difference 68588bf wireless-regdb: Update regulatory info for Syria (SY) for 2020 0dda57e wireless-regdb: Update regulatory info for Moldova (MD) on 6GHz for 2022 b19ab0b wireless-regdb: Update regulatory info for Azerbaijan (AZ) on 6GHz for 2024 f67f40d wireless-regdb: Update regulatory info for Oman (OM) bd70876 wireless-regdb: Update regulatory rules for Armenia (AM) on 2.4 and 5 GHz 6c7cbcc wireless-regdb: Permit 320 MHz bandwidth in 6 GHz band in ETSI/CEPT f9f6b30 wireless-regdb: Update regulatory rules for Austria (AT) 39b47ea wireless-regdb: Update regulatory info for Cayman Islands (KY) for 2024 3dd7ceb wireless-regdb: allow NO-INDOOR flag in db.txt 4d754a1 wireless-regdb: Update regulatory rules for Iran (IR) on both 2.4 and 5Ghz for 2021 8c8308a wireless-regdb: Update frequency range with NO-INDOOR for Oman (OM) c2f11e2 wireless-regdb: update regulatory database based on preceding changes
Antony Kolitsos [Tue, 13 May 2025 03:20:13 +0000 (20:20 -0700)]
mbedtls: update to 2.28.10
This release of Mbed TLS provides bug fixes and minor enhancements. This release includes fixes for security issues.
Mbed TLS 2.28.10 is the last release of the 2.28 LTS and won't receive bug fixes or security fixes anymore.
Users are advised to upgrade to a maintained version.
* CVE-2025-27810: Potential authentication bypass in TLS handshake [1]
* CVE-2025-27809: TLS clients should generally call mbedtls_ssl_set_hostname [2]
Andrew MacIntyre [Mon, 21 Apr 2025 13:18:40 +0000 (23:18 +1000)]
kernel: add missing symbol (BLOCK_LEGACY_AUTOLOAD)
As of kernel v5.15.179 kernel configuration fails without a
reference to BLOCK_LEGACY_AUTOLOAD, apparently as a consequence
of commit 3e9899c12d5ab3a8ce376bf4365175ea3dbaffc0 (backport of 6c0f5898836c05c6d850a750ed7940ba29e4e6c5), so add this symbol as
unset which is the same state the symbol has in the 6.6 kernel
config for 24.10.
Fixes: dfe1b5e82a35 ("kernel: Update to version 5.15.179") Signed-off-by: Andrew MacIntyre <andymac@pcug.org.au> Link: https://github.com/openwrt/openwrt/pull/18552 Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Removed the following patches because they were upstreamed:
bcm27xx/patches-5.15/950-0020-drm-probe-helper-Create-a-HPD-IRQ-event-helper-for-a.patch [0]
bcm47xx/patches-5.15/701-bgmac-reduce-max-frame-size-to-support-just-MTU-1500.patch [1]
bcm53xx/patches-5.15/700-bgmac-reduce-max-frame-size-to-support-just-MTU-1500.patch [1]
Manually adapted the following patches:
bcm27xx/patches-5.15/950-0151-usb-add-plumbing-for-updating-interrupt-endpoint-int.patch
bcm27xx/patches-5.15/950-0600-xhci-quirks-add-link-TRB-quirk-for-VL805.patch
Hauke Mehrtens [Sun, 16 Feb 2025 16:59:58 +0000 (16:59 +0000)]
kernel: Update to version 5.15.174
Manually adapted the following patches:
bcm27xx/patches-5.15/950-0156-drm-v3d-Skip-MMU-flush-if-the-device-is-currently-of.patch
bcm27xx/patches-5.15/950-0173-drm-v3d-Delete-pm_runtime-support.patch
generic/hack-5.15/780-usb-net-MeigLink_modem_support.patch
mac80211: fix compilation error for old stable kernel version
Fix compilation error for old stable version caused by
genlmsg_multicast_allns backport fix pushed middle version.
Version 5.15 version 0-169, 6.1 version 0-115, 6.6 version 0-58 have the
old genlmsg_multicast_allns version with flags variable.
Compiling backport project with these version result in a compilation
error. To handle this, introduce a backport function for the affected
kernel version.
Sat Mar 16 19:11:39 2024 daemon.info lldpd[10916]: custom TLV op replace oui 33:44:55 subtype fe
Sat Mar 16 19:11:39 2024 daemon.info lldpd[10916]: custom TLV op add oui 33:44:44 subtype e8
Sat Mar 16 19:11:39 2024 daemon.info lldpd[10916]: custom TLV op add oui 33:44:33 subtype 42
Sat Mar 16 19:11:39 2024 daemon.info lldpd[10916]: custom TLV op add oui 33:44:33 subtype 42
Sat Mar 16 19:11:39 2024 daemon.info lldpd[10916]: custom TLV op add oui 33:44:31 subtype 2c
Sat Mar 16 19:11:39 2024 daemon.info lldpd[10916]: custom TLV op add oui 33:44:31 subtype 2c
Sat Mar 16 19:11:39 2024 daemon.warn lldpcli[10915]: invalid OUI value '3322'
Sat Mar 16 19:11:39 2024 daemon.info lldpcli[10915]: an error occurred while executing last command
Sat Mar 16 19:11:39 2024 daemon.warn lldpcli[10915]: invalid OUI value '3312'
Sat Mar 16 19:11:39 2024 daemon.info lldpcli[10915]: an error occurred while executing last command
Sat Mar 16 19:11:39 2024 daemon.info lldpcli[10915]: lldpd should resume operations
( The last two TLV are invalid: their oui must be three hex bytes, comma
separated. Only the first hex byte of oui-info 5555555555 is used )
Quoting the kconfig description for CONFIG_PCPU_DEV_REFCNT:
network device refcount are using per cpu variables if this option is
set. This can be forced to N to detect underflows (with a performance
drop).
This was introduced from kernel 5.13 and was wrongly set as disabled.
Some target actually enables it but this should be always enabled unless
refcount needs to be debugged (unlikely for production images)
Enable in generic and drop the entry in every other target.
Cedric CHEDALEUX [Mon, 17 Feb 2025 09:44:36 +0000 (10:44 +0100)]
scripts/feeds: shallow clone submodules
When a feed has submodules, all its submodules are fully cloned whereas
the feed itself is shallowed. Let's be consistent and perform shallow clones
as well for the submodules.
Cedric CHEDALEUX [Mon, 17 Feb 2025 09:41:32 +0000 (10:41 +0100)]
scripts/feeds: shallow clone for specific commit update
When a feed is referenced with a specific commit (i.e. <git_url>^<sha1>),
a full clone was performed and a branch was created from the sha1
and named with the sha1. Other git clones operations are shallowed.
As Git does not support clone at a specific commit, let's first perform
a shallow clone to latest commit, then fetch the relevant commit and
finally checkout it (no more 'pseudo' branch).
It saves bandwith and significantly speeds up the feed update process.
Daniel Golle [Sat, 1 Feb 2025 05:03:50 +0000 (05:03 +0000)]
mediatek: mt7622: fix SATA on BPi-R64
A commit which made their way into Linux stable down to 5.15 broke the
SATA support on the BPi-R64.
Fix this by importing a (still pending) patch re-adding the 'syscon'
compatible to the pciesys clock-controller which also contains phy-mode
bits referenced by the ahci_mtk driver expecting to access them using
syscon_regmap_lookup_by_phandle().
Sander Vanheule [Fri, 24 Jan 2025 16:07:21 +0000 (17:07 +0100)]
realtek: Fix old compatible for HPE 1920-8G PoE
Supported devices are listed in the metadata as the first part of the
DTS compatible. This normally follows the format "vendor,device".
When updating the device name of the 180W 1920-8G PoE an underscore was
used, instead of a comma, to join the vendor and device name. This will
lead to warnings for users wanting to sysupgrade a device with an older
compatible, as the device's info does not match the one the metadata.
Fixes: 987c96e88927 ("realtek: rename hpe,1920-8g-poe to match hardware") Signed-off-by: Sander Vanheule <sander@svanheule.net>
(cherry picked from commit 6a7fa68569ac11bb2c2abb14026e8b84acd3a12f)
This patch is needed on bmips since it fixes issues with GPIOs not being
properly configured due to gpio_request_enable not being called on bcm63xx
devices. Therefore we can now drop the bcm63268 gpio function patch.
David Lutz [Tue, 5 Nov 2024 09:55:43 +0000 (10:55 +0100)]
ath79: Add support for Sophos AP15C
The Sophos AP15C uses the same hardware as the AP15, but has a reset button.
Based on:
commit 6f1efb289837 ("ath79: add support for Sophos AP100/AP55 family")
author Andrew Powers-Holmes <andrew@omnom.net>
Fri, 3 Sep 2021 15:53:57 +0200 (23:53 +1000)
committer Hauke Mehrtens <hauke@hauke-m.de>
Sat, 16 Apr 2022 16:59:29 +0200 (16:59 +0200)
Unique to AP15C:
- Reset button
- External RJ45 serial console port
Flashing instructions:
This firmware can be flashed either via a compatible Sophos SG or XG
firewall appliance, which does not require disassembling the device, or via
the U-Boot console available on the internal UART header.
To flash via XG appliance:
- Register on Sophos' website for a no-cost Home Use XG firewall license
- Download and install the XG software on a compatible PC or virtual
machine, complete initial appliance setup, and enable SSH console access
- Connect the target AP device to the XG appliance's LAN interface
- Approve the AP from the XG Web UI and wait until it shows as Active
(this can take 3-5 minutes)
- Connect to the XG appliance over SSH and access the Advanced Console
(Menu option 5, then menu option 3)
- Run `sudo awetool` and select the menu option to connect to an AP via
SSH. When prompted to enable SSH on the target AP, select Yes.
- Wait 2-3 minutes, then select the AP from the awetool menu again. This
will connect you to a root shell on the target AP.
- Copy the firmware to /tmp/openwrt.bin on the target AP via SCP/TFTP/etc
- Run `mtd -r write /tmp/openwrt.bin astaro_image`
- When complete, the access point will reboot to OpenWRT.
To flash via U-Boot serial console:
- Configure a TFTP server on your PC, and set IP address 192.168.99.8 with
netmask 255.255.255.0
- Copy the firmware .bin to the TFTP server and rename to 'uImage_AP15C'
- Open the target AP's enclosure and locate the 4-pin 3.3V UART header [4]
- Connect the AP ethernet to your PC's ethernet port
- Connect a terminal to the UART at 115200 8/N/1 as usual
- Power on the AP and press a key to cancel autoboot when prompted
- Run the following commands at the U-Boot console:
- `tftpboot`
- `cp.b $fileaddr 0x9f070000 $filesize`
- `boot`
- The access point will boot to OpenWRT.
Kyle Hendry [Sun, 17 Nov 2024 02:16:31 +0000 (18:16 -0800)]
bmips: dts: fix pinctrl error
The kernel logs the error "bcm6368_nand 10000200.nand: there is not valid
maps for state default" on boot and all nand pins show as UNCLAIMED in
sysfs pinmux-pins.
bcm6362.dtsi, bcm6368.dtsi and bcm63268.dtsi use the undocumented property
group which the driver doesn't understand. This has been documented upstream
in commit caf963efd4b0b9ff42ca12e52b8efe277264d35b.
Replacing group with pins allows the nand pins to be properly configured.
Signed-off-by: Kyle Hendry <kylehendrydev@gmail.com>
[add bcm636/bcm6368 and fix commit title] Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
(cherry picked from commit d1e9c50d06a8cb618cb85ab489cbcccaec220636)
Chris Jones [Sat, 19 Oct 2024 20:50:54 +0000 (21:50 +0100)]
ipq40xx: make GL.iNet A1300 switch functional
Set the physical switch to KEY_RFKILL, since its previous value
(KEY_SETUP) is unsupported. This should also make the KEY_RESET button
functional, by allowing the gpio-button-hotplug kmod to load.
build: Unconditionally append kmod feed for BuildBot
Recent changes to BuildBot config moved the kmods to a dedicated
directory and dropped them from the packages dir. This was needed as
both OPKG and APK gets confused if both entry are present.
To fix this, unconditionally append the kmod feed line if
CONFIG_BUILDBOT is enabled.
Michael Pratt [Fri, 24 Nov 2023 10:10:12 +0000 (05:10 -0500)]
build: ensure silent Make behavior for json scripts
Run the invocation of Make with verbosity in order to
prevent the printing of Makefile level and subtarget status.
e.g. make[3] -C target/linux val.DEFAULT_PACKAGES val.ARCH_PACKAGES
Remove piping of stderr, which is only useful when using
the "communicate" method over the "run" method,
and this script would not be written to handle a captured error anyway.
For error testing, stdout and stderr can be set to a file object
with the open() function like this:
out = open('json_out', 'w')
err = open('json_err', 'w')
Hauke Mehrtens [Fri, 22 Nov 2024 23:34:30 +0000 (00:34 +0100)]
wifi-scripts: Fix parsing of Capabilities
Fixup capabilities parsing in iw output.
In addition to the normal capabilities iw now also outputs HE MAC, HE
PHY and EHT MAC and EHT PHY capabilities. Exclude them in the parsing.
The grep returns this with mac80211-hwsim:
```
root@OpenWrt:~# iw phy phy0 info | grep 'Capabilities:'
Capabilities: 0x107e
HE PHY Capabilities: (0x02bfce0000000000000000):
EHT PHY Capabilities: (0x7c0000feffff7f01):
HE PHY Capabilities: (0x02bfce0000000000000000):
EHT PHY Capabilities: (0x7c0000feffff7f01):
HE PHY Capabilities: (0x02bf000000000000000000):
Capabilities: 0x107e
HE PHY Capabilities: (0x1cbfce0000000000000000):
EHT PHY Capabilities: (0xfc1f3ffeffff7f37):
HE PHY Capabilities: (0x1cbfce0000000000000000):
EHT PHY Capabilities: (0xfc1f3ffeffff7f37):
HE PHY Capabilities: (0x1cbf000000000000000000):
HE PHY Capabilities: (0x1cbfce0000000000000000):
EHT PHY Capabilities: (0xfefffffeffffff7f):
HE PHY Capabilities: (0x1cbfce0000000000000000):
EHT PHY Capabilities: (0xfefffffeffffff7f):
HE PHY Capabilities: (0x1cbf000000000000000000):
Capabilities: 0x107e
```
With busybox 1.36.1 the ht_cap_mask variable will be set to
-72057598332895361. With busybox 1.37.0 it will be set to -1.
Both values are wrong, after this change it will be set to 4222
(0x107E).
Huawei AP6010DN is a dual-band, dual-radio 802.11a/b/g/n 2x2 MIMO
enterprise access point with one Gigabit Ethernet port and PoE
support.
Hardware highlights:
- CPU: AR9344 SoC at 480MHz
- RAM: 128MB DDR2
- Flash: 32MB SPI-NOR
- Wi-Fi 2.4GHz: AR9344-internal radio
- Wi-Fi 5GHz: AR9580 PCIe WLAN SoC
- Ethernet: 10/100/1000 Mbps Ethernet through Atheros AR8035 PHY
- PoE: yes
- Standalone 12V/2A power input
- Serial console externally available through RJ45 port
- External watchdog: CAT706SVI (1.6s timeout)
Serial console:
9600n8 (9600 baud, no stop bits, no parity, 8 data bits)
MAC addresses:
Each device has 32 consecutive MAC addresses allocated by
the vendor, which don't overlap between devices.
This was confirmed with multiple devices with consecutive
serial numbers.
The MAC address range starts with the address on the label.
To be able to distinguish between the interfaces,
the following MAC address scheme is used:
- eth0 = label MAC
- radio0 (Wi-Fi 2.4GHz) = label MAC + 1
- radio1 (Wi-Fi 5GHz) = label MAC + 2
Installation:
0. Connect some sort of RJ45-to-USB adapter to "Console" port of the AP
1. Power up the AP
2. At prompt "Press f or F to stop Auto-Boot in 3 seconds",
do what they say.
Log in with default admin password "admin@huawei.com".
3. Boot the OpenWrt initramfs from TFTP using the hidden script "run ramboot".
Replace IP address as needed:
4. Optional but recommended as the factory firmware cannot be downloaded publicly:
Back up contents of "firmware" partition using the web interface or ssh:
5. Run sysupgrade using sysupgrade image. OpenWrt
shall boot from flash afterwards.
Return to factory firmware (using firmware upgrade package downloaded from non-public Huawei website):
1. Start a TFTP server in the directory where
the firmware upgrade package is located
2. Boot to u-boot as described above
3. Install firmware upgrade package and format the config partitions:
> update system FatAP6X10XN_SOMEVERSION.bin
> format_fs
Return to factory firmware (from previously created backup):
1. Copy over the firmware partition backup to /tmp,
for example using scp
2. Use sysupgrade with force to restore the backup:
sysupgrade -F huawei_ap6010dn_fw_backup.bin
3. Boot AP to U-Boot as described above
Quirks and known issues:
- The stock firmware has a semi dual boot concept where the primary
kernel uses a squashfs as root partition and the secondary kernel uses
an initramfs. This dual boot concept is circumvented on purpose to gain
more flash space and since the stock firmware's flash layout isn't
compatible with mtdsplit.
- The external watchdog's timeout of 1.6s is very hard to satisfy
during bootup. This is why the GPIO15 pin connected to the watchdog input
is configured directly in the LZMA loader to output the AHB_CLK/2 signal
which keeps the watchdog happy until the wdt-gpio kernel driver takes
over. Because it would also take too long to read the whole kernel image
from flash, the uImage header only includes the loader which then reads
the kernel image from flash after GPIO15 is configured.
Hauke Mehrtens [Sun, 17 Nov 2024 21:15:21 +0000 (22:15 +0100)]
kernel: bump 5.15 to 5.15.173
Removed because they are upstream:
generic/backport-5.15/430-v6.3-udf-Allocate-name-buffer-in-directory-iterator-on-he.patch
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-5.15.y&id=5ea4b73e268bf9e1d26271943f2f7c4517253395
Fix genlmsg_multicast_allns() build error on kernel 6.6.59.
Based on kernel patch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit?h=v6.6.59&id=e0f83d268974dab0361d11904dfc9acec53f96a6
Hauke Mehrtens [Mon, 4 Nov 2024 22:45:25 +0000 (23:45 +0100)]
kernel: bump 5.15 to 5.15.170
Removed because they are upstream:
generic/backport-5.15/777-netfilter-xtables-fix-typo-causing-some-targets-to-not-load-on-IPv6.patch
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=linux-5.15.y&id=90baa455aa7e099152898cfa5eb3928d6152da12
Hauke Mehrtens [Mon, 4 Nov 2024 21:12:03 +0000 (22:12 +0100)]
kernel: bump 5.15 to 5.15.169
Added patch:
generic/backport-5.15/430-v6.3-udf-Allocate-name-buffer-in-directory-iterator-on-he.patch
This patch fixes the following compile warning:
```
CC [M] fs/udf/namei.o
fs/udf/namei.c: In function 'udf_rename':
fs/udf/namei.c:878:1: error: the frame size of 1144 bytes is larger than 1024 bytes [-Werror=frame-larger-than=]
878 | }
| ^
cc1: all warnings being treated as errors
make[7]: *** [scripts/Makefile.build:289: fs/udf/namei.o] Error 1
```
Hauke Mehrtens [Tue, 29 Oct 2024 20:48:11 +0000 (21:48 +0100)]
kernel: bump 5.15 to 5.15.168
Manually adapted:
bcm27xx/patches-5.15/950-0166-xhci-Use-more-event-ring-segment-table-entries.patch
I replaced it with the version from main branch which applies more easily
bcm27xx/patches-5.15/950-0600-xhci-quirks-add-link-TRB-quirk-for-VL805.patch
bcm27xx/patches-5.15/950-0605-xhci-refactor-out-TRBS_PER_SEGMENT-define-in-runtime.patch
bcm27xx/patches-5.15/950-0606-usb-xhci-add-VLI_TRB_CACHE_BUG-quirk.patch
bcm27xx/patches-5.15/950-0717-usb-xhci-add-a-quirk-for-Superspeed-bulk-OUT-transfe.patch
bcm53xx/patches-5.15/180-usb-xhci-add-support-for-performing-fake-doorbell.patch
generic/hack-5.15/645-netfilter-connmark-introduce-set-dscpmark.patch
Added patch:
generic/backport-5.15/777-netfilter-xtables-fix-typo-causing-some-targets-to-not-load-on-IPv6.patch
This fixes a bug intoduced in this version.
Adapt kernel configuration for newly added option
CONFIG_PROC_MEM_ALWAYS_FORCE.
Rafal Boni [Thu, 31 Oct 2024 22:54:21 +0000 (18:54 -0400)]
ipq806x: Fix default MAC addresses on MR42 by moving to nvmem-layout
Partial, single-target update extracted from https://github.com/openwrt/openwrt/commit/d264d3a6
The previous `mac-address-increment` is deprecated, and in particular on
this target means that the kernel is unable to read the MAC address,
causing the system to boot with a new random MAC address each time.
Fixes: https://github.com/openwrt/openwrt/issues/13992 Signed-off-by: Rosen Penev <rosenp@gmail.com>
[rafal.boni@gmail.com: single-target-specific backport from larger change] Signed-off-by: Rafal Boni <rafal.boni@gmail.com> Link: https://github.com/openwrt/openwrt/pull/16858 Signed-off-by: Robert Marko <robimarko@gmail.com>
Lech Perczak [Thu, 24 Oct 2024 18:51:00 +0000 (20:51 +0200)]
ath79: Really fix 5GHz on QCA9886 variant of ZTE MF286
Fix in commit 25eead21c5ab ("ath79: fix 5GHz on QCA9886 variant of ZTE MF286")
was incomplete. A user of such variant popped up, and in the boot log
after installation, we discovered that QCA9886 expects different
pre-calibration data size, than the older QCA9880 variant:
ath10k_pci 0000:00:00.0: qca9888 hw2.0 target 0x01000000 chip_id 0x00000000 sub 0000:0000
ath10k_pci 0000:00:00.0: kconfig debug 0 debugfs 1 tracing 0 dfs 1 testmode 0
ath10k_pci 0000:00:00.0: firmware ver 10.4b-ct-9888-fW-13-5ae337bb1 api 5 features mfp,peer-flow-ctrl,txstatus-noack,wmi-10.x-CT,ratemask-CT,regdump-CT,txrate-CT,flush-all-CT,pingpong-CT,ch-regs-CT,nop-CT,set-special-CT,tx-rc-CT,cust-stats-CT,txrate2-CT,beacon-cb-CT,wmi-block-ack-CT,wmi-bcn-rc-CT crc32 59e741e7
ath10k_pci 0000:00:00.0: invalid calibration data length in nvmem-cell 'pre-calibration': 2116 != 12064
ath10k_pci 0000:00:00.0: Loading BDF type 0
ath10k_pci 0000:00:00.0: failed to fetch board data for bus=pci,vendor=168c,device=0056,subsystem-vendor=0000,subsystem-device=0000 from ath10k/QCA9888/hw2.0/board-2.bin
ath10k_pci 0000:00:00.0: failed to fetch board-2.bin or board.bin from ath10k/QCA9888/hw2.0
ath10k_pci 0000:00:00.0: failed to fetch board file: -12
ath10k_pci 0000:00:00.0: could not probe fw (-12)
Explicitly define a pre-calibration nvmem-cell for this variant, and use
it instead of the calibration one, which is shorter.
Rafal Boni [Thu, 24 Oct 2024 15:40:53 +0000 (11:40 -0400)]
ipq806x: Fix default MAC addresses on MR52 by moving to `nvmem-layout`
Partial, single-target update extracted from https://github.com/openwrt/openwrt/commit/d264d3a6
The previous `mac-address-increment` is deprecated, and in particular on
this target means that the kernel is unable to read the MAC address,
causing the system to boot with a new random MAC address each time.
Fixes: https://github.com/openwrt/openwrt/issues/15238 Signed-off-by: Rosen Penev <rosenp@gmail.com>
[rafal.boni@gmail.com: single-target-specific backport from larger change] Signed-off-by: Rafal Boni <rafal.boni@gmail.com> Link: https://github.com/openwrt/openwrt/pull/16774 Signed-off-by: Robert Marko <robimarko@gmail.com>
Rafal Boni [Thu, 24 Oct 2024 07:06:22 +0000 (03:06 -0400)]
ipq806x: Revert "ipq806x: swap lan leds for Meraki MR52"
This reverts commit ec8f647d168fa8f3b1eedd9b5fe665f793f3a659, as with the
current kernel version, the change actually causes the same bug it once
may have fixed -- that is, the leds are now again reversed.
I suspect this was due to a switch to a newer kernel version between when
the patch was submitted and now reversing the order of the interfaces, so
that eth0 / the LAN interface is also the interface used for PoE, and eth1
/ the WAN interface is the non-PoE interface.
ath79: add support for MikroTik RouterBOARD 750 r2 (hEX lite)
This patch adds support for the MikroTik RouterBOARD 750 r2, marketed as
hEX lite, a small indoor router with 5x 10/100 Mbps Ethernet ports, one
with PoE in. The device was already supported by the ar71xx target.
See https://mikrotik.com/product/RB750r2 for more details.
Not working:
- Serial port (already not working in ar71xx)
Flashing:
TFTP boot initramfs image and then perform sysupgrade. Only the
"Internet" port will ask for an initramfs image. Follow common
MikroTik procedure as in https://openwrt.org/toh/mikrotik/common.
projects / thirdparty / openwrt.git / log
