openwrt/openwrt.git
8 weeks agomediatek: mt7988: fix peripheral SPI busses
Daniel Golle [Fri, 1 Nov 2024 03:37:51 +0000 (03:37 +0000)]
mediatek: mt7988: fix peripheral SPI busses

The clocks for SPI busses were named wrongly which resulted in the
spi-mt65xx driver not requesting them. This has apparently been
worked around by marking the clocks required for SPI0 which is used
for SPI-NOR and SPI-NAND flash chips as critical.
Fix the device tree for all 3 generic SPI host controllers and no
longer mark clocks as critical.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
(cherry picked from commit 2b173ab730897f53539cab39e87e2b4638cb482a)

8 weeks agoopenwrt-keyring: add OpenWrt 24.10 release build usign key
Petr Štetiar [Fri, 1 Nov 2024 06:25:07 +0000 (06:25 +0000)]
openwrt-keyring: add OpenWrt 24.10 release build usign key

 fbae29d730f8 usign: add OpenWrt 24.10 release build public key

Signed-off-by: Petr Štetiar <ynezz@true.cz>
(cherry picked from commit a535cfc09e1ced5785d65bee0a0116a71585128e)

8 weeks agoOpenWrt v24.10: set branch defaults
Christian Marangi [Thu, 31 Oct 2024 21:15:37 +0000 (22:15 +0100)]
OpenWrt v24.10: set branch defaults

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
8 weeks agoath79: Really fix 5GHz on QCA9886 variant of ZTE MF286
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.

Fixes: 25eead21c5ab ("ath79: fix 5GHz on QCA9886 variant of ZTE MF286")
Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16809
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
8 weeks agobcm27xx: update to latest RPi patches
Álvaro Fernández Rojas [Thu, 31 Oct 2024 09:09:44 +0000 (10:09 +0100)]
bcm27xx: update to latest RPi patches

The patches were generated from the RPi repo with the following command:
git format-patch v6.6.58..rpi-6.6.y

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
8 weeks agobcm27xx-gpu-fw: update to latest version
Álvaro Fernández Rojas [Wed, 30 Oct 2024 23:25:41 +0000 (00:25 +0100)]
bcm27xx-gpu-fw: update to latest version

Use release tar instead of downloading separate files.

Full changelog: https://github.com/raspberrypi/firmware/compare/1.20240424...1.20241008

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
8 weeks agoramips: m4r v4: fix typo
Rosen Penev [Fri, 25 Oct 2024 22:39:30 +0000 (15:39 -0700)]
ramips: m4r v4: fix typo

It should be debounce-interval, as with the others.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16802
Signed-off-by: Robert Marko <robimarko@gmail.com>
8 weeks agoCI: labeler: add bcm27xx-utils
Álvaro Fernández Rojas [Thu, 31 Oct 2024 07:52:16 +0000 (08:52 +0100)]
CI: labeler: add bcm27xx-utils

bcm27xx-utils can only be built for bcm27xx target.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
8 weeks agobcm27xx-utils: update to latest version
Álvaro Fernández Rojas [Wed, 30 Oct 2024 12:49:02 +0000 (13:49 +0100)]
bcm27xx-utils: update to latest version

Fixes build with GCC 14.

Full changelog: https://github.com/raspberrypi/utils/compare/451b9881b72cb994c102724b5a7d9b93f97dc315...6a2a6becebbc38fde34a94386457ac8210f9119b

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
8 weeks agoipq806x: Revert "ipq806x: swap lan leds for Meraki MR52"
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.

Signed-off-by: Rafal Boni <rafal.boni@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16779
Signed-off-by: Robert Marko <robimarko@gmail.com>
8 weeks agoramips: ignore errors from trx utility
Christian Marangi [Wed, 30 Oct 2024 13:07:22 +0000 (14:07 +0100)]
ramips: ignore errors from trx utility

Ignore errors from trx utility to handle case with image too big.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
8 weeks agoimage: ignore errors from more commands
Christian Marangi [Wed, 30 Oct 2024 13:05:10 +0000 (14:05 +0100)]
image: ignore errors from more commands

Ignore errors in more image commands to handle case where the image is
too big and check-image validation fails.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
8 weeks agoimagebuilder: fix copying of missing kernel/libc
Paul Spooren [Wed, 30 Oct 2024 12:41:23 +0000 (13:41 +0100)]
imagebuilder: fix copying of missing kernel/libc

Those packages were not copied due to OPKG using an underscore while APK
uses dashes. Remove that char to copy kernel/libc for either APK/OPKG.

Signed-off-by: Paul Spooren <mail@aparcar.org>
8 weeks agoimage: ignore errors from mktplinkfw command
Christian Marangi [Wed, 30 Oct 2024 10:21:58 +0000 (11:21 +0100)]
image: ignore errors from mktplinkfw command

Rework tplink-v2-header and tplink-v2-image Build define to ignore error
if mktplinkfw2 errors out.

This is to handle situation when the image is too big and can't be
generated or prev check-image calls deleted the source file as it's too
big.

This aligns to the pattern used by tplink-v1-image.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
8 weeks agobuild: correctly ignore Initramfs image on error
Christian Marangi [Wed, 30 Oct 2024 10:19:33 +0000 (11:19 +0100)]
build: correctly ignore Initramfs image on error

Correctly ignore Initramfs image copy on error. This follows the pattern with
sysupgrade image where an image might fail as it's too big or the
generation command fails and there is nothing to copy to the bin
directory.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
8 weeks agomxs: drop removal of firewall4 package
Christian Marangi [Wed, 30 Oct 2024 09:25:25 +0000 (10:25 +0100)]
mxs: drop removal of firewall4 package

Drop removal of firewall4 package for I2SE Duckbill device.

With OPKG the firewall4 package was installed anyway as it's a
dependency of luci-app-firewall and was silently installed again later
in such condition. Drop it to fix support for APK.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
8 weeks agomvebu: cortexa9: drop removal of firewall4 package
Christian Marangi [Wed, 30 Oct 2024 09:23:10 +0000 (10:23 +0100)]
mvebu: cortexa9: drop removal of firewall4 package

Drop removal of firewall4 package for Synology DS213j device.

With OPKG the firewall4 package was installed anyway as it's a
dependency of luci-app-firewall and was silently installed again later
in such condition. Drop it to fix support for APK.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
8 weeks agobuild: Honor DEFAULT_VARIANT for APK packages
Paul Spooren [Wed, 30 Oct 2024 08:28:30 +0000 (09:28 +0100)]
build: Honor DEFAULT_VARIANT for APK packages

Previously APK would complain as it wasn't sure which package to
install by default when multiple packages would provide the same name.

Now, give the package a higher provider priority to make APK
automatically select the "default" package.

Signed-off-by: Paul Spooren <mail@aparcar.org>
8 weeks agobuild: don't include kernel/libc in package index
Paul Spooren [Tue, 29 Oct 2024 10:09:24 +0000 (11:09 +0100)]
build: don't include kernel/libc in package index

The same that is done in `ipkg-make-index.sh` should happen with APK.
If the pseudo packages, only added to add dependency constraints, are
added to the index, APK happily "upgrades" them and installs updated
kmods, too. However, the Kernel itself is never installed via a regular
package.

Fixes: https://github.com/openwrt/openwrt/issues/16808
Signed-off-by: Paul Spooren <mail@aparcar.org>
8 weeks agokernel: mtdsplit_minor: accept bootimage filename
John Thomson [Wed, 16 Oct 2024 02:51:22 +0000 (12:51 +1000)]
kernel: mtdsplit_minor: accept bootimage filename

RouterBOOT v7 on NOR devices no longer accepts the YAFFS kernel ELF
method of booting. It will accept an NPK image named bootimage.
Adjust mtdsplit_minor to accept this second possible boot file name.
Use the conservative value of 127 for YAFFS max name length (used when
YAFFS compiled with unicode support) vs 255.

Signed-off-by: John Thomson <git@johnthomson.fastmail.com.au>
Acked-by: Thibaut VARENE <hacks@slashdirt.org>
Link: https://github.com/openwrt/openwrt/pull/16780
Signed-off-by: Robert Marko <robimarko@gmail.com>
8 weeks agokernel: mtdsplit_minor: return 0 if not fatal
John Thomson [Tue, 15 Oct 2024 21:13:25 +0000 (07:13 +1000)]
kernel: mtdsplit_minor: return 0 if not fatal

Introduced with Linux 6.7, in commit:
5c2f7727d437 ("mtd: mtdpart: check for subpartitions parsing result"),
when a parser returns an error, this will be passed up, and
consequently, all parent mtd partitions get torn down.

Adjust the MiNOR mtdsplit driver to only return an error if there is a
critical problem in reading from the mtd device or allocating memory.
Otherwise return 0 to indicate that no partitions were found.
Also add logging to indicate what went wrong.

This mtdsplit parser makes a very limited check of the first YAFFS
header. For example, this will not match expectations when initially booting
an initramfs image with OEM on MTD.

Signed-off-by: John Thomson <git@johnthomson.fastmail.com.au>
Acked-by: Thibaut VARENE <hacks@slashdirt.org>
Link: https://github.com/openwrt/openwrt/pull/16780
Signed-off-by: Robert Marko <robimarko@gmail.com>
8 weeks agoramips: linkit: replace driver with gpio-hog
Rosen Penev [Tue, 22 Oct 2024 00:14:41 +0000 (17:14 -0700)]
ramips: linkit: replace driver with gpio-hog

It does the same thing minus a few dmesg prints.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16788
Signed-off-by: Robert Marko <robimarko@gmail.com>
8 weeks agoramips: Fix WiFi on Sercomm NA502 and NA502s
Andreas Böhler [Mon, 28 Oct 2024 17:52:02 +0000 (18:52 +0100)]
ramips: Fix WiFi on Sercomm NA502 and NA502s

The WiFi chips are actually on PCIe1 and PCIe2, PCIe0 is empty. Fix the
assignment so that WiFi works properly again.

Signed-off-by: Andreas Böhler <dev@aboehler.at>
Link: https://github.com/openwrt/openwrt/pull/16807
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
8 weeks agobcm27xx-utils: fix PKG_MIRROR_HASH
Álvaro Fernández Rojas [Tue, 29 Oct 2024 14:30:09 +0000 (15:30 +0100)]
bcm27xx-utils: fix PKG_MIRROR_HASH

The mirror hash has changed after 8009342.

Fixes: 8009342f43c5 ("bcm27xx-utils: fix version for APK")
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
8 weeks agogeneric: backport upstream r8169 irq patch
Álvaro Fernández Rojas [Tue, 29 Oct 2024 08:58:57 +0000 (09:58 +0100)]
generic: backport upstream r8169 irq patch

This commit backports an additional patch from linux v6.12 release.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
2 months agoimagebuilder: always add local packages.adb
Paul Spooren [Mon, 28 Oct 2024 22:20:57 +0000 (23:20 +0100)]
imagebuilder: always add local packages.adb

It doesn't matter if `IB_STANDALONE` is enabled or not, local packages
must always be considered.

Signed-off-by: Paul Spooren <mail@aparcar.org>
2 months agopackage: fix key creation for SDK
Paul Spooren [Mon, 28 Oct 2024 21:33:48 +0000 (22:33 +0100)]
package: fix key creation for SDK

Prior to this commit keys would only be generated if `make` is called
alone, but not for something like `make package/busybox/compile`.

The exact reasons are in the depth of make magic, so this is sheer luck!

Signed-off-by: Paul Spooren <mail@aparcar.org>
2 months agoinclude/package-pack: drop version from provides
Paul Spooren [Sun, 27 Oct 2024 15:36:33 +0000 (16:36 +0100)]
include/package-pack: drop version from provides

Adding a version to the provides causes it to conflict with other
packages that provides the same package, further details are available
here: https://gitlab.alpinelinux.org/alpine/apk-tools/-/blob/master/doc/apk-package.5.scd#L199

This was intitally done, if I remember correctly, to support depending
on the specific kernel modules. Due to patches to APK, versions
containing hashes work now, too, so this is no longer required.

Only add the version to packages that define an ABI version since other
packages depend against the package name plus ABI version.

While at it, format the now rather complex call.

Fixes: https://github.com/openwrt/openwrt/issues/16795
Signed-off-by: Paul Spooren <mail@aparcar.org>
2 months agoarmsr: don't limit kmods to armsr armv8 subtarget
Christian Marangi [Sun, 27 Oct 2024 09:02:55 +0000 (10:02 +0100)]
armsr: don't limit kmods to armsr armv8 subtarget

Drop limitation on depending on only armv8 for armsr target as those
module should support both 32 and 64 bits systems.

Only thunderx-net actually require 64 bit systems.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agodsl-vrx200-firmware-xdsl: fix PKG_VERSION for APK
Paul Spooren [Fri, 12 Apr 2024 17:01:57 +0000 (19:01 +0200)]
dsl-vrx200-firmware-xdsl: fix PKG_VERSION for APK

The version was a mix of strings, hex numbers and semantic numbers.

Switch the PKG_VERSION to something digestible by APK and introduce
PKG_SOURCE_VERSION to handle the actual filename.

While at it, drop the redundant PKG_B_NAME which was the same as
PKG_NAME anyway.

Signed-off-by: Paul Spooren <mail@aparcar.org>
2 months agolayerscape: drop fmc and fmc-eth-config package
Christian Marangi [Sat, 26 Oct 2024 21:17:46 +0000 (23:17 +0200)]
layerscape: drop fmc and fmc-eth-config package

Drop fmc and fmc-eth-config package as they were never actually
submitted to openwrt mainline and they don't exist around.

They are probably part of NXP SDK and were added due to copy-paste
errors.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agolayerscape: set firmware package to skip installation
Christian Marangi [Sat, 26 Oct 2024 21:16:21 +0000 (23:16 +0200)]
layerscape: set firmware package to skip installation

Add the '~' prefix to package that needs to skip installation as they
are meta-package just to download and compile firmware package for the
final firmware.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agobuild: introduce support to declare skip package
Christian Marangi [Sat, 26 Oct 2024 20:57:46 +0000 (22:57 +0200)]
build: introduce support to declare skip package

It seems some target started declaring package in DEVICE_PACKAGES just
to call InstallDev and generate binary for the image firmware.

This is very much used by layerscape target where trusted-firmware-a and
dependency are called for final image generation.

This is problematic for APK since it's more sensible to non exisiting
package.

To handle this, introduce a prefix '~' for a package that will signal to
build the package but not install it in the final image.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoomnia-mcutool: fix version for APK
Christian Marangi [Sat, 26 Oct 2024 10:29:32 +0000 (12:29 +0200)]
omnia-mcutool: fix version for APK

Refactor version of omnia-mcutool for APK.

Switch to git clone and use hash instead of converting 0.3-rc3 to 0.3.3.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoadb: fix version for APK
Christian Marangi [Sat, 26 Oct 2024 09:50:21 +0000 (11:50 +0200)]
adb: fix version for APK

Refactor version of adb for APK.

Use HASH instead of rc tag.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoapk: switch to wget url backend for HOST tool
Christian Marangi [Fri, 25 Oct 2024 20:35:02 +0000 (22:35 +0200)]
apk: switch to wget url backend for HOST tool

Switch APK to wget url backend for HOST tool to handle a problem with
libressl.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoramips: mt7621: drop kmod-i2c-ralink from ASUS RP-AC56
Christian Marangi [Fri, 25 Oct 2024 12:56:20 +0000 (14:56 +0200)]
ramips: mt7621: drop kmod-i2c-ralink from ASUS RP-AC56

Drop kmod-i2c-ralink from ASUS RP-AC56 as it was wrongly added. Such
kmod is not supported on mt7621 as i2c is handled by the mediatek driver
and not bay the ralink downstream one.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agobcm27xx-utils: fix version for APK
Christian Marangi [Fri, 25 Oct 2024 10:27:28 +0000 (12:27 +0200)]
bcm27xx-utils: fix version for APK

Refactor version of bcm27xx-utils for APK.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agomsx: fix wrong device package list for I2SE Duckbill
Christian Marangi [Fri, 25 Oct 2024 09:56:45 +0000 (11:56 +0200)]
msx: fix wrong device package list for I2SE Duckbill

Fix wrong package device list that is trying to remove deprecated
packages. Replace with new variant where possible.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoadb-enablemodem: fix version for APK
Christian Marangi [Fri, 25 Oct 2024 08:25:10 +0000 (10:25 +0200)]
adb-enablemodem: fix version for APK

Refactor version of adb-enablemodem for APK.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoapk: add patch adding support for ARM Big-Endian arch
Christian Marangi [Thu, 24 Oct 2024 23:12:35 +0000 (01:12 +0200)]
apk: add patch adding support for ARM Big-Endian arch

Add patch adding support for ARM Big-Endian arch like Intel XScale
IXP4xx SoC.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoipq40xx: drop ipq-wifi-teltonika_rutx from Teltonika RUTX50
Christian Marangi [Thu, 24 Oct 2024 22:47:07 +0000 (00:47 +0200)]
ipq40xx: drop ipq-wifi-teltonika_rutx from Teltonika RUTX50

Drop ipq-wifi-teltonika_rutx from Teltonika RUTX50, the board file was
merged upstream but the ipq package was never dropped from
DEVICE_PACKAGES list.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agomac80211: reintroduce rt61/rt73 support
Christian Marangi [Thu, 24 Oct 2024 22:12:05 +0000 (00:12 +0200)]
mac80211: reintroduce rt61/rt73 support

Reintroduce rt61/rt73 support as they looks to be pretty standard and
currently required by Gemini or Xway-legacy targets.

Notice that they are b/g card.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agobcm53xx: drop kmod-ledtrig-default-on and kmod-ledtrig-netdev
Christian Marangi [Thu, 24 Oct 2024 21:46:44 +0000 (23:46 +0200)]
bcm53xx: drop kmod-ledtrig-default-on and kmod-ledtrig-netdev

Drop kmod-ledtrig-default-on and kmod-ledtrig-netdev as the kmod were
dropped and are now enabled by default.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoath79: replace ancient kmod-ledtrig-usbdev with kmod-usb-ledtrig-usbport
Christian Marangi [Thu, 24 Oct 2024 21:37:24 +0000 (23:37 +0200)]
ath79: replace ancient kmod-ledtrig-usbdev with kmod-usb-ledtrig-usbport

Replace ancient kmod-ledtrig-usbdev dropped from 2016 with
kmod-usb-ledtrig-usbport upstream.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agobcm27xx-gpu-fw: fix version for APK
Christian Marangi [Thu, 24 Oct 2024 21:23:03 +0000 (23:23 +0200)]
bcm27xx-gpu-fw: fix version for APK

Refactor version of bcm27xx-gpu-fw for APK.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoimagebuilder: complete support for local signing keys
Christian Marangi [Thu, 24 Oct 2024 18:45:01 +0000 (20:45 +0200)]
imagebuilder: complete support for local signing keys

Complete support for local signing keys for APK.

A local key will be always generated, mkndx is always called with
--allow-untrusted as it needs to replace the sign key with the new local
one.

With CONFIG_SIGNATURE_CHECK the local index is signed with the local
key. Local public key is added with the ADD_LOCAL_KEY option.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoimagebuilder: actually support IB from buildbot
Christian Marangi [Thu, 24 Oct 2024 18:36:43 +0000 (20:36 +0200)]
imagebuilder: actually support IB from buildbot

ImageBuilder compiled by buildbot doesn't have any package in the
packages directory. Package needs to be downloaded instead.

This works by calling update to the package manage to download the
remove index and download the file.

Fix missing support for this with APK, by configuring the
--repositories-file option and calling the APK update.

Also move the apk add --initdb to package_index.

If CONFIG_SIGNATURE_CHECK is not enabled, the signature is not checked.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoimagebuilder: correctly export PACKAGE_DIR and PACKAGE_DIR_ALL
Christian Marangi [Thu, 24 Oct 2024 18:32:42 +0000 (20:32 +0200)]
imagebuilder: correctly export PACKAGE_DIR and PACKAGE_DIR_ALL

Correctly export PACKAGE_DIR and PACKAGE_DIR_ALL so that they won't be
reset on internal call of rules.mk

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoimagebuilder: copy apk keys with CONFIG_SIGNATURE_CHECK
Christian Marangi [Thu, 24 Oct 2024 18:29:52 +0000 (20:29 +0200)]
imagebuilder: copy apk keys with CONFIG_SIGNATURE_CHECK

With CONFIG_SIGNATURE_CHECK enabled, copy keys following OPKG pattern.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agobuild: detach apk repository handling from rootfs.mk
Christian Marangi [Thu, 24 Oct 2024 18:25:06 +0000 (20:25 +0200)]
build: detach apk repository handling from rootfs.mk

To better support imagebuilder declaring --repositories-file on calling
apk macro, detach this and --repository from rootfs.mk macro and move it
to package Makefile and image.mk where they are used to permit a more
generic usage.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agobuild: permit overwrite PACKAGE_DIR and PACKAGE_DIR_ALL
Christian Marangi [Thu, 24 Oct 2024 18:13:07 +0000 (20:13 +0200)]
build: permit overwrite PACKAGE_DIR and PACKAGE_DIR_ALL

Permit to overwrite PACKAGE_DIR and PACKAGE_DIR_ALL variables in
rules.mk.

This is to handle a special case with the ImageBuilder where these
variable are overwrite.

The main problem is that any include calling rules.mk again (example
image.mk) will set these variables again dropping the modified value.

To keep the modified value, set the PACKAGE_DIR and PACKAGE_DIR_ALL only
if not already set. This permits the ImageBuilder to use custom
directory instead of the default one defined in rules.mk.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agocypress-firmware: fix version for APK
Paul Spooren [Fri, 23 Aug 2024 21:46:52 +0000 (23:46 +0200)]
cypress-firmware: fix version for APK

Refactor version for cypress-firmware for APK.

Signed-off-by: Paul Spooren <mail@aparcar.org>
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoapk: bump to latest HEAD 2024-10-28
Christian Marangi [Mon, 28 Oct 2024 23:03:39 +0000 (00:03 +0100)]
apk: bump to latest HEAD 2024-10-28

Bump APK to latest HEAD 2024-10-28.

- Minor fixed to index inherit
- Fixup to adb_w_copy

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoairoha: an7581: replace TRNG patch with upstream version
Christian Marangi [Mon, 28 Oct 2024 13:01:52 +0000 (14:01 +0100)]
airoha: an7581: replace TRNG patch with upstream version

Replace TRNG patch with upstream version.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoxdp-tools: fix compilation when bpftool is installed
Til Kaiser [Sun, 27 Oct 2024 13:02:14 +0000 (14:02 +0100)]
xdp-tools: fix compilation when bpftool is installed

Currently, xdp-tools doesn't compile on build systems where
bpftool is installed because additional tools and BPF programs
will be compiled then, which results in build errors.

This commit disables the compilation of those bpftool-dependent tools.

Signed-off-by: Til Kaiser <mail@tk154.de>
Link: https://github.com/openwrt/openwrt/pull/16787
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agodns320l-mcu: bump to 2024-10-27
Zoltan HERPAI [Sun, 27 Oct 2024 17:36:12 +0000 (17:36 +0000)]
dns320l-mcu: bump to 2024-10-27

Improve support for power button handling.

d9a2878 - Use /sys/class/gpio/mcu_power for monitoring the MCU power line
02b6005 - Use "halt" instead of "shutdown"

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
2 months agokirkwood: dns320l: export mcu_power GPIO
Zoltan HERPAI [Sun, 27 Oct 2024 17:35:11 +0000 (17:35 +0000)]
kirkwood: dns320l: export mcu_power GPIO

This GPIO is pulled down by the onboard MCU when the power button
is pressed for 5 seconds, indicating a user-initiated shutdown.

Refresh patches at the same time.

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
2 months agorealtek: ethernet: Improve SMI polling configuration based on DTS
Markus Stockhausen [Sun, 15 Sep 2024 17:01:41 +0000 (13:01 -0400)]
realtek: ethernet: Improve SMI polling configuration based on DTS

Although Zyxel XGS1210 devices are not yet officially supported there
are several patches floating around to enable them. This is a very imporant
one because it fixes a SMI misconfiguration. In the known DTS the SFP+
port settings are set as follows.

  phy26: ethernet-phy@26 {
    compatible = "ethernet-phy-ieee802.3-c45";
    phy-is-integrated;
    reg = <26>;
    sds = < 8 >;
  };

  phy27: ethernet-phy@27 {
    compatible = "ethernet-phy-ieee802.3-c45";
    phy-is-integrated;
    reg = <27>;
    sds = < 9 >;
  };

So these are PHYs linked to an internal SerDes. During initialization
rtl838x_mdio_init() generates smi_bus=0 & smi_addr=27/28 for these ports.
Although this seems like a valid configuration integrated PHYs attached
to an SerDes do not have an SMI bus. Later on the mdio reset wrongly feeds
the SMI registers and as a result the PHYs on SMI bus 0 do not work.

Without patch (loaded with rtk network on & initramfs):

...
mdio_bus mdio-bus: MDIO device at address 0 is missing.
mdio_bus mdio-bus: MDIO device at address 1 is missing.
mdio_bus mdio-bus: MDIO device at address 2 is missing.
mdio_bus mdio-bus: MDIO device at address 3 is missing.
mdio_bus mdio-bus: MDIO device at address 4 is missing.
mdio_bus mdio-bus: MDIO device at address 5 is missing.
mdio_bus mdio-bus: MDIO device at address 6 is missing.
mdio_bus mdio-bus: MDIO device at address 7 is missing.
...
rtl83xx-switch ... : no phy at 0
rtl83xx-switch ... : failed to connect to PHY: -ENODEV
rtl83xx-switch ... : error -19 setting up PHY for tree 0, switch 0, port 0
rtl83xx-switch ... : no phy at 1
rtl83xx-switch ... : failed to connect to PHY: -ENODEV
rtl83xx-switch ... : error -19 setting up PHY for tree 0, switch 0, port 1
...

With patch (loaded with rtk network on & initramfs):

...
rtl83xx-switch ... : PHY [mdio-bus:00] driver [REALTEK RTL8218D] (irq=POLL)
rtl83xx-switch ... : PHY [mdio-bus:01] driver [REALTEK RTL8218D] (irq=POLL)
...

Signed-off-by: Birger Koblitz <git@birger-koblitz.de>
Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/16457
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agorealtek: remove wrong SMI bus from XGS1250
Markus Stockhausen [Fri, 18 Oct 2024 06:20:20 +0000 (02:20 -0400)]
realtek: remove wrong SMI bus from XGS1250

The RTL930x have only 4 SMI busses (0-3) and the XGS1250 SFP port ist
directly managed. Remove the wrong configuration in the dts.

Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/16457
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agorealtek: phy: fix RTL8218D detection
Markus Stockhausen [Sun, 22 Sep 2024 17:36:14 +0000 (13:36 -0400)]
realtek: phy: fix RTL8218D detection

Currently RTL8218D detection works for a range of devices. That can lead to
false positives. E.g. RTL8218B or RTL8214FC are covered by the detection mask
as well. That is wrong. Nail detection down to the real RTL8218D phy id.

Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/16457
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agorealtek: phy: simplify RTL8214C detection
Markus Stockhausen [Fri, 18 Oct 2024 07:00:18 +0000 (03:00 -0400)]
realtek: phy: simplify RTL8214C detection

The detection of the RTL8214C is a little complicated. Make it easier.

Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/16457
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agorealtek: phy: proper RTL8218B, RTL8214FC, RTL8214FB detection
Markus Stockhausen [Sun, 22 Sep 2024 17:19:22 +0000 (13:19 -0400)]
realtek: phy: proper RTL8218B, RTL8214FC, RTL8214FB detection

Three PHYs share the same identifier. Until now we simply assume
the type depending of the bus address it is attached to. Make it
better and check the chip mode register instead.

The kernel will either detect by id/mask or by match_phy_device().
Remove the unneeded settings.

Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/16457
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agorealtek: phy: adapt raw page for RTL839X
Markus Stockhausen [Sun, 22 Sep 2024 16:46:18 +0000 (12:46 -0400)]
realtek: phy: adapt raw page for RTL839X

The number of phy pages differ between RTL838X and RTL839X. Make that
clear and adapt the existing defines.

Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/16457
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agorealtek: dsa: increase RTL839x max phy page to 8191
Markus Stockhausen [Sun, 22 Sep 2024 16:03:43 +0000 (12:03 -0400)]
realtek: dsa: increase RTL839x max phy page to 8191

According to the specs the RTL839x provides up to 8192 phy pages.
Especially the "raw" page 8191 is used for different initialization
tasks. Increase the limit.

Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/16457
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agorealtek: dsa: allow USXGMII mode
Markus Stockhausen [Fri, 18 Oct 2024 06:14:02 +0000 (02:14 -0400)]
realtek: dsa: allow USXGMII mode

RTL930x devices need the USXGMII mode. This is a final leftover
from the 6.6 conversion.

Signed-off-by: Markus Stockhausen <markus.stockhausen@gmx.de>
Link: https://github.com/openwrt/openwrt/pull/16457
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agoRevert "scripts/download.pl: read file natively"
Christian Marangi [Sun, 27 Oct 2024 14:58:16 +0000 (15:58 +0100)]
Revert "scripts/download.pl: read file natively"

This reverts commit 65560de58de7a3cde8f960849e3c7c640b9456d8.

Seems to broke CI and require more testing.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agomediatek: mt7988: add DT entries for WED
Daniel Golle [Sun, 27 Oct 2024 14:24:47 +0000 (14:24 +0000)]
mediatek: mt7988: add DT entries for WED

Add memory regions and devices used for wireless offloading to the
device tree for MT7988.

This allows using WED on devices with MT7988 SoC and MT7995E, MT7996E or
MT7992E wireless controllers.

Devices with 4 GiB of RAM (or more) will still need ajustments to avoid
running out of swiotlb entries.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
2 months agoramips: modernize pwm driver
Rosen Penev [Wed, 9 Oct 2024 20:44:56 +0000 (13:44 -0700)]
ramips: modernize pwm driver

Add more devm usage.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16649
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 months agorockchip: Add LED aliases for FriendlyElec NanoPi R3S
Kevin Zhang [Sun, 27 Oct 2024 05:13:47 +0000 (01:13 -0400)]
rockchip: Add LED aliases for FriendlyElec NanoPi R3S

This commit adds OpenWRT's LED aliases to the board DTS
for showing system status on the NanoPi R3S.

Signed-off-by: Kevin Zhang <kevin@kevinzhang.me>
Link: https://github.com/openwrt/openwrt/pull/16738
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 months agorockchip: Add support for RK3566 FriendlyElec NanoPi R3S
Kevin Zhang [Sun, 27 Oct 2024 05:13:05 +0000 (01:13 -0400)]
rockchip: Add support for RK3566 FriendlyElec NanoPi R3S

This commit adds support for the FriendlyElec NanoPi R3S.

CPU: Rockchip RK3566, Quad-core Cortex-A55
RAM: 2GB LPDDR4X
Ethernet: GMAC RTL8211F GbE, PCIe R8111H GbE
USB3.0 Host: Type-A x1
Storage: MicroSD Slot x 1, and optional on-board 32GB eMMC
Debug Serial Port: 3.3V TTL, 3-pin 2.54mm pitch connector, 1500000 bauds
LED: LED x 3
RTC: One low-power RTC, supports backup battery input

Both GbE controllers are working (WAN eth0, LAN eth1).
Appropriate LAN/WAN interface assignments and MAC address generation.
All three LEDs are working.
USB appears to be working and has been tested with mass storage.

Installation - microSD:

-Uncompress the OpenWRT sysupgrade.img.gz
-Write image to microSD card using dd or similar tool

Installation - eMMC:

-Boot from microSD
-Uncompress the OpenWRT sysupgrade.img.gz
-Flash to eMMC : dd if=x.img of=/dev/mmcblk0
-sync
-Remove microSD card
-Reboot

Signed-off-by: Kevin Zhang <kevin@kevinzhang.me>
Link: https://github.com/openwrt/openwrt/pull/16738
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 months agouboot-rockchip: add support for NanoPi R3S
Kevin Zhang [Sun, 27 Oct 2024 05:12:01 +0000 (01:12 -0400)]
uboot-rockchip: add support for NanoPi R3S

This commit adds u-boot support for the NanoPi R3S.

Signed-off-by: Kevin Zhang <kevin@kevinzhang.me>
Link: https://github.com/openwrt/openwrt/pull/16738
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 months agoramips: use platform_get_irq
Rosen Penev [Tue, 22 Oct 2024 23:21:46 +0000 (16:21 -0700)]
ramips: use platform_get_irq

No need for irq_of_parse_and_map since this is in _probe.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16771
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoscripts/download.pl: read file natively
Matteo Croce [Sat, 28 Sep 2024 23:01:01 +0000 (01:01 +0200)]
scripts/download.pl: read file natively

Read files natively instead of execing /bin/cat.

Signed-off-by: Matteo Croce <teknoraver@meta.com>
Link: https://github.com/openwrt/openwrt/pull/16784
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoxdp-tools: re-add vlan_hdr struct
Til Kaiser [Sun, 27 Oct 2024 13:05:50 +0000 (14:05 +0100)]
xdp-tools: re-add vlan_hdr struct

Drop the patch to re-add the vlan_hdr struct
because it is required now to build xdp-filter.

Signed-off-by: Til Kaiser <mail@tk154.de>
Link: https://github.com/openwrt/openwrt/pull/16798
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agokernel: add missing CONFIG_DM_AUDIT
Marius Dinu [Sat, 26 Oct 2024 17:10:55 +0000 (20:10 +0300)]
kernel: add missing CONFIG_DM_AUDIT

The option shows up when kmod-dm and CONFIG_KERNEL_AUDIT are selected.

Signed-off-by: Marius Dinu <m95d+git@psihoexpert.ro>
Link: https://github.com/openwrt/openwrt/pull/16793
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 months agokernel: add MultiPath TCP diag kernel module
Daniel Golle [Sat, 26 Oct 2024 00:35:29 +0000 (01:35 +0100)]
kernel: add MultiPath TCP diag kernel module

Add INET diag socket monitoring interface module for MultiPath TCP.
It allows using native Linux socket diagnostic tools such as 'ss' for
Multipath TCP connections.

Co-authored-by: Rodrigo B. de Sousa Martins <rodrigo.sousa.577@gmail.com>
Signed-off-by: sKy King <29267720+sKyissKy@users.noreply.github.com>
Link: https://github.com/openwrt/openwrt/pull/12884
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
2 months agokernel: enable Multi-Path TCP for !SMALL_FLASH targets
Daniel Golle [Fri, 25 Oct 2024 17:15:41 +0000 (18:15 +0100)]
kernel: enable Multi-Path TCP for !SMALL_FLASH targets

Expose Kernel's CONFIG_MPTCP option and enable it by default for
!SMALL_FLASH targets.

The idea behind enabling it by default is to allow users of the binary
distribution to make use of MPTCP tunneling for link aggregation.

Using MPTCP for link aggregation is an often discussed topic in the
forum and there is even a whole OpenWrt fork (MPTCPRouter) just for that.

Enabling the kernel-side of the story by default will allow using MPTCP
on vanilla OpenWrt without having to build anything from source.

See also https://openwrt.org/docs/guide-user/network/mptcp

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
2 months agogeneric: mtk_wed: fix path of MT7988 WO firmware
Daniel Golle [Sat, 26 Oct 2024 14:18:50 +0000 (15:18 +0100)]
generic: mtk_wed: fix path of MT7988 WO firmware

commit eee3c695f3 ("linux-firmware: add offloading firmware for MT7988")
added mt7988_wo_{0,1}.bin in the 'mediatek/mt7988' directory while driver
currently expects the files in the 'mediatek' directory.

Import pending patch which changes the path in the driver header now
that the firmware has been added.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
2 months agoairoha: Clarify subtarget compatibility
Andreas Gnau [Fri, 25 Oct 2024 16:49:48 +0000 (18:49 +0200)]
airoha: Clarify subtarget compatibility

Clarify compatibility of the two subtargets with different SoCs to
reduce confusion. The added SoC names only differ in small details such
as features enabled (PoN vs DSL for example).

Signed-off-by: Andreas Gnau <andreas.gnau@iopsys.eu>
Link: https://github.com/openwrt/openwrt/pull/16785
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoramips: rt3883: convert devices to interrupt-driven gpio-keys
Shiji Yang [Wed, 23 Oct 2024 16:09:36 +0000 (00:09 +0800)]
ramips: rt3883: convert devices to interrupt-driven gpio-keys

This converts all RT3662 and RT3883 devices to use interrupt based
gpio-keys instead of gpio-keys-polled. The poll-interval will be
removed.

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Link: https://github.com/openwrt/openwrt/pull/16764
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agoramips: rt305x: convert devices to interrupt-driven gpio-keys
Shiji Yang [Wed, 23 Oct 2024 16:09:19 +0000 (00:09 +0800)]
ramips: rt305x: convert devices to interrupt-driven gpio-keys

This converts all RT305x and RT5350 devices to use interrupt based
gpio-keys instead of gpio-keys-polled. The poll-interval will be
removed.

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Link: https://github.com/openwrt/openwrt/pull/16764
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agoramips: rt2880: convert devices to interrupt-driven gpio-keys
Shiji Yang [Wed, 23 Oct 2024 16:08:25 +0000 (00:08 +0800)]
ramips: rt2880: convert devices to interrupt-driven gpio-keys

This converts all RT2880 devices to use interrupt based gpio-keys
instead of gpio-keys-polled. The poll-interval will be removed.

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Link: https://github.com/openwrt/openwrt/pull/16764
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agoramips: mt76x8: convert Minew G1-C to interrupt-driven gpio-keys
Shiji Yang [Wed, 23 Oct 2024 16:07:53 +0000 (00:07 +0800)]
ramips: mt76x8: convert Minew G1-C to interrupt-driven gpio-keys

All other MT76x8 devices have already been migrated to gpio-keys.

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Link: https://github.com/openwrt/openwrt/pull/16764
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agoramips: dts: add the missing interrupt properties for GPIO nodes
Shiji Yang [Wed, 23 Oct 2024 15:09:57 +0000 (23:09 +0800)]
ramips: dts: add the missing interrupt properties for GPIO nodes

The Ralink GPIO driver supports irqchip function. Hence we need to
add "interrupt-parent" and "interrupt-controller" properties to make
sure it works properly. It is worth noting that all GPIO devices
share the same interrupt line.

Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Link: https://github.com/openwrt/openwrt/pull/16764
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agoramips: ralink-gpio: use irqchip helpers to register driver
Shiji Yang [Wed, 23 Oct 2024 13:56:23 +0000 (21:56 +0800)]
ramips: ralink-gpio: use irqchip helpers to register driver

The gpiolib has already introduced a general GPIO irqchip framework
to initialize the GPIO irqchip[1]. This patch will make use of it
to simplify the legacy Ralink GPIO driver codes. This patch also
includes some code readability improvements.

[1] 1425052097b5 ("gpio: add IRQ chip helpers in gpiolib")
Signed-off-by: Shiji Yang <yangshiji66@qq.com>
Link: https://github.com/openwrt/openwrt/pull/16764
Signed-off-by: Robert Marko <robimarko@gmail.com>
2 months agombedtls: update to 3.6.2
Hauke Mehrtens [Wed, 23 Oct 2024 22:03:13 +0000 (00:03 +0200)]
mbedtls: update to 3.6.2

Fixes the following security problem:
* CVE-2024-49195: Fix a buffer underrun in mbedtls_pk_write_key_der()
  when called on an opaque key, MBEDTLS_USE_PSA_CRYPTO is enabled, and
  the output buffer is smaller than the actual output. Fix a related
  buffer underrun in mbedtls_pk_write_key_pem() when called on an opaque
  RSA key, MBEDTLS_USE_PSA_CRYPTO is enabled and MBEDTLS_MPI_MAX_SIZE is
  smaller than needed for a 4096-bit RSA key.

Link: https://github.com/openwrt/openwrt/pull/16768
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 months agoinclude/package-pack: Set missing Maintainer and URL for APK
Christian Marangi [Thu, 24 Oct 2024 10:46:47 +0000 (12:46 +0200)]
include/package-pack: Set missing Maintainer and URL for APK

Set missing Maintainer and URL info for .apk creation.

Fixes: d788ab376f85 ("build: add APK package build capabilities")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoinclude/package-pack: remove whitespace for Description info for APK
Christian Marangi [Thu, 24 Oct 2024 10:45:36 +0000 (12:45 +0200)]
include/package-pack: remove whitespace for Description info for APK

Remove whitespace for Description info for .apk creation

Fixes: b6bbc76c0b2a ("include/package-pack: set missing Description on .apk creation")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agompc85xx: move CONFIG_MTD_SPLIT_FIRMWARE up
Rosen Penev [Thu, 17 Oct 2024 23:27:02 +0000 (16:27 -0700)]
mpc85xx: move CONFIG_MTD_SPLIT_FIRMWARE up

It's shared by all targets. Unlike the other shared symbols, it's not
removed by kernel_oldconfig.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16756
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 months agoapm821xx: move CONFIG_GENERIC_PHY up
Rosen Penev [Thu, 17 Oct 2024 23:48:20 +0000 (16:48 -0700)]
apm821xx: move CONFIG_GENERIC_PHY up

It's shared by both subtargets.

Signed-off-by: Rosen Penev <rosenp@gmail.com>
Link: https://github.com/openwrt/openwrt/pull/16754
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 months agokernel: bump 6.6 to 6.6.58
John Audia [Tue, 22 Oct 2024 19:59:20 +0000 (15:59 -0400)]
kernel: bump 6.6 to 6.6.58

Changelog: https://cdn.kernel.org/pub/linux/kernel/v6.x/ChangeLog-6.6.58

All patches automatically rebased.

Build system: x86/64
Build-tested: x86/64/AMD Cezanne, flogic/glinet_gl-mt6000, flogic/xiaomi_redmi-router-ax6000-ubootmod, ramips/tplink_archer-a6-v3
Run-tested: x86/64/AMD Cezanne, flogic/glinet_gl-mt6000, flogic/xiaomi_redmi-router-ax6000-ubootmod, ramips/tplink_archer-a6-v3

Signed-off-by: John Audia <therealgraysky@proton.me>
Link: https://github.com/openwrt/openwrt/pull/16758
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 months agomediatek: add wireless offloading firmware to MT7988 boards
Daniel Golle [Wed, 23 Oct 2024 12:38:55 +0000 (13:38 +0100)]
mediatek: add wireless offloading firmware to MT7988 boards

Install mt7988-wo-firmware by default on boards based on MT7988.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
2 months agolinux-firmware: add offloading firmware for MT7988
Daniel Golle [Wed, 23 Oct 2024 12:36:29 +0000 (13:36 +0100)]
linux-firmware: add offloading firmware for MT7988

Package wireless offloading firmware for MediaTek MT7988.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
2 months agoairoha: rename subtarget from en7581 to an7581
Christian Marangi [Wed, 23 Oct 2024 10:19:03 +0000 (12:19 +0200)]
airoha: rename subtarget from en7581 to an7581

The Airoha EN7581 got renamed to AN7581 due to move from Econet to
Airoha.

To save on compatibility, use both compatible for the device.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoairoha: en7581: refresh and fix PWM patch
Christian Marangi [Wed, 23 Oct 2024 10:17:40 +0000 (12:17 +0200)]
airoha: en7581: refresh and fix PWM patch

Refresh and fix PWM patch with new revision proposed upstream.

Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agoinclude/package-pack: set missing Description on .apk creation
Christian Marangi [Tue, 22 Oct 2024 22:10:02 +0000 (00:10 +0200)]
include/package-pack: set missing Description on .apk creation

Set missing description info on .apk creation. This was probably a TODO
that wasn't notice when the final implementation was pushed.

Fixes: d788ab376f85 ("build: add APK package build capabilities")
Signed-off-by: Christian Marangi <ansuelsmth@gmail.com>
2 months agokernel: Reorder config files
Hauke Mehrtens [Sun, 20 Oct 2024 13:31:23 +0000 (15:31 +0200)]
kernel: Reorder config files

Reorder the kernel configuration files.

This was done uisng:
./scripts/kconfig-reorder.sh

Link: https://github.com/openwrt/openwrt/pull/16743
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 months agomediatek: Remove double CONFIG_HARDEN_BRANCH_PREDICTOR
Hauke Mehrtens [Sun, 20 Oct 2024 13:38:57 +0000 (15:38 +0200)]
mediatek: Remove double CONFIG_HARDEN_BRANCH_PREDICTOR

CONFIG_HARDEN_BRANCH_PREDICTOR option is included two times in this
configuration file. Remove one definition. On arm32 SoC it should not be
needed.

Fixes: 54f9744c820a ("treewide: disable spectre mitigation on unaffected Arm32 targets")
Link: https://github.com/openwrt/openwrt/pull/16743
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
2 months agoscripts: Add script to reorder kernel config-* files
Hauke Mehrtens [Mon, 21 Oct 2024 22:02:28 +0000 (00:02 +0200)]
scripts: Add script to reorder kernel config-* files

This script will reorder the content of all config-* files in the target
folder. It will also remove duplicates. It will not remove options
already defined in the generic configuration.

Link: https://github.com/openwrt/openwrt/pull/16743
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>