aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* kernel: 5.15: add missing symbolRobert Marko2022-04-301-0/+1
| | | | | | CONFIG_QCOM_SPMI_ADC_TM5 gets uncovered while building IPQ40xx with 5.15. Signed-off-by: Robert Marko <robert.marko@sartura.hr>
* kernel: bump 5.10 to 5.10.113John Audia2022-04-3015-51/+51
| | | | | | | | | | All patches automatically rebased. Build system: x86_64 Build-tested: bcm2711/RPi4B, mt7622/RT3200 Run-tested: bcm2711/RPi4B, mt7622/RT3200 Signed-off-by: John Audia <graysky@archlinux.us>
* mediatek: build ubnt-ledbar as a moduleChuanhong Guo2022-04-304-4/+18
| | | | | | | | The config for LEDS_UBNT_LEDBAR doesn't stay in mt7629 kconfig because of its I2C dependency. Build it as a module and let buildroot handle this config option instead. Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: mt7629: disable auxadc in kconfigChuanhong Guo2022-04-301-0/+1
| | | | | | manually edit kconfig to disable MT6577_AUXADC Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: mt7623: disable various input devicesChuanhong Guo2022-04-301-45/+1
| | | | | | | | Disable support for joysticks, micee and tablets. There's no actual driver selected in kconfig, and including kernel support is just a waste of space. Besides that, I believe nobody wants these on a router. Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: mt7622: add irq for spi-nor controllerChuanhong Guo2022-04-301-0/+10
| | | | | | | Save some CPU from unnecessary polling and make SPI flash reading a tiny bit faster. Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: remove left-overs from Linux 5.10Daniel Golle2022-04-3049-4508/+0
| | | | | | | | Remove patches and configuration for Linux 5.10 which have been left in the tree despite the target having been switched to Linux 5.15. Fixes: c283defa88 ("mediatek: switch to 5.15") Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* mediatek: enable multi-core schedulingDaniel Golle2022-04-301-0/+1
| | | | | | | | Set CONFIG_SCHED_MC in config-5.15 to have make the scheduler aware of shared caches. Reported-by: Rui Salvaterra <rsalvaterra@gmail.com> Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* mediatek: refresh patches for Linux 5.15Daniel Golle2022-04-3022-147/+53
| | | | | | Run `make target/linux/refresh` to align patches with OpenWrt style. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* mediatek: backport recent DTS improvements for MT7622 SoCDaniel Golle2022-04-303-0/+360
| | | | | | | | | | | Daniel Golle: 190-arm64-dts-mediatek-mt7622-fix-GICv2-range.patch Rui Salvaterra: 191-arm64-dts-mt7622-specify-the-L2-cache-topology.patch 192-arm64-dts-mt7622-specify-the-number-of-DMA-requests.patch Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* kernel: don't parse FIT partition on NAND-backed mtdblockDaniel Golle2022-04-301-4/+3
| | | | | | | | | | | ubiblock devices should be used on NAND flash to store the uImage.FIT in case the bootloader supports that -- otherwise only rootfs is stored in UBI while the uImage.FIT contains only the kernel and dtb. Hence there is no need to enable parsing partitions on NAND mtdblock devices, it is even responsible for the ugly warning on-opening of the mtdblock device now. Just don't do it. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* kernel: backport patch to relax mtdblock NAND warningDaniel Golle2022-04-301-0/+82
| | | | | | | Instead of warning loudly about mtdblock devices being created, rather just warn if they are actually used. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* mediatek: mt7629: disable mtk thermalChuanhong Guo2022-04-291-0/+1
| | | | | | Manually edit kconfig to disable CONFIG_MTK_THERMAL Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: mt7623: disable ecc engine supportChuanhong Guo2022-04-291-1/+1
| | | | | | There's no board using nand on this target atm. Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: mt7623 enable arch-timer with a patchChuanhong Guo2022-04-292-5/+27
| | | | | | | | | Last attempt on this has a typo and doen't work. It seems that this is a common problem occurring on every kernel bump, so let's enforce arch timer support for mt7623 with a patch instead. Fixes: 9a22943eb2 ("mediatek: 5.15: re-enable arch timer on MT7623 as well") Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* ipq40xx: 5.10: fix ar40xx driverNick Hainke2022-04-291-0/+3
| | | | | | | | | | | | | | | | | | | | | | | | This commit is completely based on the work of adron-s: https://github.com/openwrt/openwrt/pull/4721#issuecomment-1101108651 The commit fixes the data corruption on TX packets. Packets are transmitted, but their contents are replaced with zeros. This error is caused by the lack of guard (50 ms) intervals between calibration phases. This error is treated by adding mdelay(50) to the calibration function code. In the original qca-ssda code [0], these mdelays were existing, but in the ar41xx.c they are gone. Tested on: - Fritz!Box 4040 - Fritz!Box 7530 - Mikrotik SXTsq 5AC - ZyXEL NBG6617 - [0] https://git.codelinaro.org/clo/qsdk/oss/lklm/qca-ssdk/-/blob/NHSS.QSDK.11.4/src/init/ssdk_init.c#L2072 Suggested-by: Serhii Serhieiev <adron@mstnt.com> Reviewed-by: Robert Marko <robimarko@gmail.com> Signed-off-by: Nick Hainke <vincent@systemli.org>
* mediatek: switch to 5.15Chuanhong Guo2022-04-281-2/+1
| | | | | | | | the SPI-NAND driver switch breaks dts compatibility. It's too much work to backport all ECC framework support to 5.10 so let's switch the target to 5.15 instead. Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: enable snfi for mt7629Chuanhong Guo2022-04-281-9/+15
| | | | Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: restore support for Fidelix FM35X1GAChuanhong Guo2022-04-281-0/+122
| | | | | | This patch was dropped when the spi-nand driver got replaced last time. Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: add an upstreamed spi-nand driverChuanhong Guo2022-04-2816-351/+3446
| | | | | | | This patch implements the spi-nand controller driver as an ECC-capable spi-mem controller to use the upstream SPI-NAND driver. Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: spi-nand: check for 1 byte BBM onlyChuanhong Guo2022-04-281-0/+20
| | | | | | | | the OOB layout in MTK SNFI uses the 2nd byte, and anything using OOB will make the block a bad-block in spi-nand driver. Hack it for now. We need a proper solution upstream. Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: v5.15: backport spi-mem ecc supportChuanhong Guo2022-04-2810-0/+928
| | | | Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* mediatek: drop spi-nand driver from mtkChuanhong Guo2022-04-2812-4036/+0
| | | | Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* ucode: reorder BuildPackage callsJo-Philipp Wich2022-04-281-1/+1
| | | | | | | | | Ensure that the libucode recipe is processed before the ucode one in order to reliably encode the ABI version into ucode's libucode dependency. Fixes: #9788 Ref: https://forum.openwrt.org/t/fw4-wont-start-after-upgrade/126308 Signed-off-by: Jo-Philipp Wich <jo@mein.io>
* ath79: add support for TP-Link EAP265 HDSander Vanheule2022-04-272-1/+4
| | | | | | | | The EAP265 HD is a rebadged EAP245v3, so images are compatible with both devices. Link: https://fccid.io/TE7EAP265HD/Letter/6-Request-for-FCC-Change-ID-4823578.pdf Signed-off-by: Sander Vanheule <sander@svanheule.net>
* firmware-utils: bump to git HEADSander Vanheule2022-04-271-3/+3
| | | | | | | | | Includes image support for new TP-Link devices: ddc3e00e314d tplink-safeloader: add TP-Link EAP265 HD support ceea1a7fe56e tplink-safeloader: add TP-Link Deco M4R v1 and v2 support Signed-off-by: Sander Vanheule <sander@svanheule.net>
* iwinfo: update to latest HEADDavid Bauer2022-04-271-3/+3
| | | | | | dc6847e iwinfo: nl80211: omit A-hwmode on non-5GHz hardware Signed-off-by: David Bauer <mail@david-bauer.net>
* nat46: update to git HEADHans Dedecker2022-04-261-3/+3
| | | | | | | | | 95ca1c3 nat46-core: ignore IPv4 options when translating packets 39778c2 add a module argument to ignore TOS translate for IPv4 9a36ee1 add a module argument to ignore TOS translate for IPv4 79190a8 add a module argument to ignore TOS translate for IPv4 Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
* uboot-envtools: add WS-AP3825i configDavid Bauer2022-04-261-0/+4
| | | | | | | Add configuration to use uboot-envtools with the Extreme Networks WS-AP3825i. Signed-off-by: David Bauer <mail@david-bauer.net>
* mpc85xx: set WS-AP3825i mac-address in preinitDavid Bauer2022-04-261-0/+12
| | | | | | | | | The bootloader does seem to not correctly patch in the MAC address for eth0 / eth1 in some cases. While the root cause is not known, manually applying the MAC-Address in preinit does not hurt. Reported-by: Tom Herbers <freifunk@tomherbers.de> Signed-off-by: David Bauer <mail@david-bauer.net>
* mpc85xx: define reset-delay for WS-AP3825i eth PHYDavid Bauer2022-04-261-0/+4
| | | | | | | | | | | | The WS-AP3825i uses Atheros PHYs which according to the datasheet require the reset to be asserted for at least 1 ms. This fixes broken eth1 upon soft-reboot. eth0 is no affected, as the ifup / ifdown cycle in preinit prevents this issue from happening when the system is ready. Reported-by: Tom Herbers <freifunk@tomherbers.de> Signed-off-by: David Bauer <mail@david-bauer.net>
* dnsmasq: add logfacility file to jail mountsDaniel Golle2022-04-251-2/+6
| | | | | | | | If logfacility is a path to a file it needs to be r/w mounted in the sandbox as well for dnsmasq to work. Reported-by: @iointerrupt Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* iwinfo: update to latest HEADDavid Bauer2022-04-241-3/+3
| | | | | | | a479b9b devices: remove whitespace 562d015 iwinfo: nl80211: fix hwmode parsing for multi-band NICs Signed-off-by: David Bauer <mail@david-bauer.net>
* ath79: ubnt: drop swconfig on ac-{lite,lr,mesh}Martin Weinelt2022-04-231-0/+3
| | | | | | These don't have switches that could be configured using swconfig. Signed-off-by: Martin Weinelt <hexa@darmstadt.ccc.de>
* ramips: add support for TP-Link RE650 v2Marcin Gordziejewski2022-04-234-1/+209
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | TP-Link RE650 v2 is largely similar to v1 that is already supported by OpenWrt. Notable differences is differnt SPI Flash - 8 MB instead of 16 MB (from cFeon instead of Winbond) and a different configuration of PCIE connections to wifi chips. Otherwise it's largely the same product as v1 Hardware specification: - SoC 880 MHz - MediaTek MT7621AT - 128 MB of DDR3 RAM - 8 MB - cFeon QH64A-104HIP - 4T4R 2.4 GHz - MediaTek MT7615E - 4T4R 5 GHz - MediaTek MT7615E - 1x 1 Gbps Ethernet - MT7621AT integrated - 7x LEDs (Power, 2G, 5G, WPS(x2), Lan(x2)) - 4x buttons (Reset, Power, WPS, LED) - UART pinout - GND, RX, TX, labeled in the middle of the PCB, requires soldering because they're not through holes. Serial console @ 57600,8n1 Flash instructions: Upload openwrt-ramips-mt7621-tplink_re650-v2-squashfs-factory.bin from the RE650 web interface. TFTP recovery to stock firmware: I didn't try recovering back to the stock firmware, however, if there is such process for other RExxx devices, it seems like it could be similar here. Signed-off-by: Marcin Gordziejewski <openwrt@flicksfix.com>
* firmware-utils: bump to git HEADHauke Mehrtens2022-04-231-3/+3
| | | | | | 05fd700 tplink-safeloader: TP-Link RE650 v2 support Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
* ramips: add support for YunCore AX820/HWAP-AX820Clemens Hopfer2022-04-235-1/+164
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | There are two versions which are identical apart from the enclosure: YunCore AX820: indoor ceiling mount AP with integrated antennas YunCore HWAP-AX820: outdoor enclosure with external (N) connectors Hardware specs: SoC: MediaTek MT7621DAT Flash: 16 MiB SPI NOR RAM: 128MiB (DDR3, integrated) WiFi: MT7905DAN+MT7975DN 2.4/5GHz 2T2R 802.11ax Ethernet: 10/100/1000 Mbps x2 (WAN/PoE+LAN) LED: Status (green) Button: Reset Power: 802.11af/at PoE; DC 12V,1A Antennas: AX820(indoor): 4dBi internal; HWAP-AX820(outdoor): external Flash instructions: The "OpenWRT support" version of the AX820 comes with a LEDE-based firmware with proprietary MTK drivers and a luci webinterface and ssh accessible under 192.168.1.1 on LAN; user root, no password. The sysupgrade.bin can be flashed using luci or sysupgrade via ssh, you will have to force the upgrade due to a different factory name. Remember: Do *not* preserve factory configuration! MAC addresses as used by OEM firmware: use address source 2g 44:D1:FA:*:0b Factory 0x0004 (label) 5g 46:D1:FA:*:0b LAA of 2g lan 44:D1:FA:*:0c Factory 0xe000 wan 44:D1:FA:*:0d Factory 0xe000 + 1 The wan MAC can also be found in 0xe006 but is not used by OEM dtb. Due to different MAC handling in mt76 the LAA derived from lan is used for 2g to prevent duplicate MACs when creating multiple interfaces. Signed-off-by: Clemens Hopfer <openwrt@wireloss.net>
* kernel: bump 5.15 to 5.15.35Rui Salvaterra2022-04-233-10/+8
| | | | | | Patches automatically rebased. Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
* kernel: bump 5.10 to 5.10.112John Audia2022-04-2310-26/+24
| | | | | | | | | | | | | Manually rebased: ath79/patches-5.10/901-phy-mdio-bitbang-prevent-rescheduling-during-command.patch All other patches automatically rebased. Build system: x86_64 Build-tested: bcm2711/RPi4B, mt7622/RT3200 Run-tested: bcm2711/RPi4B, mt7622/RT3200 Signed-off-by: John Audia <graysky@archlinux.us>
* bcm63xx: fix description fix name caseHuangbin Zhan2022-04-231-1/+1
| | | | | | The `Description` should be capital. Signed-off-by: Huangbin Zhan <zhanhb88@gmail.com>
* ath79: ZTE MF286R: add comgt-ncm to DEVICE_PACKAGESLech Perczak2022-04-231-1/+2
| | | | | | | | | | | When adding support to the router's built-in modem, this required package was omitted, because it was already enabled in the image configuration in use for testing, and this went unnoticed. In result, the modem still isn't fully supported in official images. As it is the primary WAN interface, add the missing package. Fixes: e02fb42c53ba ("comgt: support ZTE MF286R modem") Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
* mt76: update to the latest versionFelix Fietkau2022-04-231-3/+3
| | | | | | a666d5637bc3 mt76: fix tx status related use-after-free race on station removal Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ucode: fix PKG_MIRROR_HASHJo-Philipp Wich2022-04-221-1/+1
| | | | | Fixes: 0400774a10 ("ucode: update to latest Git HEAD") Signed-off-by: Jo-Philipp Wich <jo@mein.io>
* firewall4: update to latest Git HEADJo-Philipp Wich2022-04-221-3/+3
| | | | | | | | fc83d46 ruleset: set auto-merge directive for interval sets 9bce873 fw4: fix skipping invalid ipset entries 425ea8a fw4: fix applying zone flags for source bound rules Signed-off-by: Jo-Philipp Wich <jo@mein.io>
* ucode: update to latest Git HEADJo-Philipp Wich2022-04-221-3/+3
| | | | | | | e14b099 syntax: implement support for ES6 template literals 111cf06 vm: stop executing bytecode on return of nested calls Signed-off-by: Jo-Philipp Wich <jo@mein.io>
* mt76: update to the latest versionFelix Fietkau2022-04-201-3/+3
| | | | | | | | | | | | | | | | d4937118ec4f mt76: fix rx reordering with non explicit / psmp ack policy f50672518bcc mt76: do not attempt to reorder received 802.3 packets without agg session 75e046e3c9ae mt76: fix antenna config missing in 6G cap 64ede85f14b5 mt76: mt7915: remove SCS feature db0a67582028 mt76: mt7915: make read-only array ppet16_ppet8_ru3_ru0 static const 68a2f68d5c01 mt76: mt7921: make read-only array ppet16_ppet8_ru3_ru0 static const 1e7b77294e97 mt76: mt7921: fix kernel crash at mt7921_pci_remove 795830a99a6c mt76: mt7915: add debugfs knob for RF registers read/write 60a8ec12a8eb mt76: mt7603: move spin_lock_bh() to spin_lock() 5ec7a2eac867 mt76: mt7915: disable RX_HDR_TRANS_SHORT eecbb4992073 mt76: fix encap offload ethernet type check Signed-off-by: Felix Fietkau <nbd@nbd.name>
* kernel: update FIT partition parser to new property nameDaniel Golle2022-04-201-1/+1
| | | | | | | | The commit "uboot-mediatek: replace patch with accepted commit" changed the name of the boot configuration property from 'bootconf' to 'u-boot,bootconf'. Reflect this change in the FIT partition parser. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* uboot-mediatek: replace patch with accepted commitDaniel Golle2022-04-201-34/+10
| | | | | | | | Replace pending patch with version accepted upstream. Other than in the first suggested version, the new property is now called 'u-boot,bootconf' instead of 'bootconf'. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* kernel: x86: remove an upstreamed patchRui Salvaterra2022-04-202-100/+0
| | | | | | | | | 012-pcengines-apu2-detect-apuv4-board.patch is upstream since Linux 5.5 [1]. [1] https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/commit/?h=v5.5&id=3d00da1de3ea36ba44f4a7ba76c8c8b16f98204b Acked-by: Petr Štetiar <ynezz@true.cz> Signed-off-by: Rui Salvaterra <rsalvaterra@gmail.com>
* wolfssl: fix compilation with /dev/cryptoEneas U de Queiroz2022-04-201-0/+19
| | | | | | This is trivial fix of a duplicate definition of 'int ret'. Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
generated by cgit v1.2.3 (git 2.25.1) at 2025-02-13 06:15:30 +0000