aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* uml: build target again via the BuildbotsPaul Spooren2021-11-021-1/+1
| | | | | | | | | | The UML target been outdated for a long time. Instead of just carrying unmaintained code we should build it again and allow people and CIs to use it for testing. This commit removes the `source-only` feature which disables building. Signed-off-by: Paul Spooren <mail@aparcar.org>
* uml: drop 32Bit host build supportPaul Spooren2021-11-023-153/+4
| | | | | | | | This step drops support host build systems other than x86/64 to allow two Kernel configuration in parallel. With this commit the setup follow the config style of all other targets. Signed-off-by: Paul Spooren <mail@aparcar.org>
* uml: update README to usable examplesPaul Spooren2021-11-022-46/+51
| | | | | | | The current `uml` README is terribly outdated and non of the examples work by default. Fix that and while at it convert it to Markdown. Signed-off-by: Paul Spooren <mail@aparcar.org>
* tools/libressl: update to 3.4.1Rosen Penev2021-11-021-2/+2
| | | | Signed-off-by: Rosen Penev <rosenp@gmail.com>
* build: fix ldconfig executable error in pythonDamien Mascord2021-11-021-2/+1
| | | | | | | | | | | | | | | | | The empty executable is causing problems with meson builds, due to the error: OSError: [Errno 8] Exec format error: 'ldconfig' This patch changes the empty ldconfig stub to symlink to /bin/true to work around this issue. Fixes: FS#4117 Fixes: 3bd31cc4d2ff ("tools/meson: update to 0.60.0") Signed-off-by: Damien Mascord <tusker@tusker.org> Tested-by: Aleksander Jan Bajkowski <olek2@wp.pl> # Tested on Debian 11 Tested-By: Lucian Cristian <lucian.cristian@gmail.com> Tested-By: Baptiste Jonglez <git@bitsofnetworks.org> Cc: Rosen Penev <rosenp@gmail.com>
* qosify: add missing dependencyFelix Fietkau2021-11-021-1/+1
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* include/bpf.mk: fix compile for big-endian targetsFelix Fietkau2021-11-021-2/+3
| | | | | | llvm-opt and llc need endian flags in the target as well Signed-off-by: Felix Fietkau <nbd@nbd.name>
* bpf-headers: unset PKG_CONFIG_PATHFelix Fietkau2021-11-021-0/+2
| | | | | | | This fixes an issue where the kernel would pick up an incompatible target libyaml for building host tools Signed-off-by: Felix Fietkau <nbd@nbd.name>
* include/bpf.mk: fix typoFelix Fietkau2021-11-021-1/+1
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* iproute2: Avoid unnecessary package rebuildsEneas U de Queiroz2021-11-011-1/+3
| | | | | | | | | | | | | | | | | | Build the tc-mod-iptables before the tc-tiny and tc-full packages. This avoids unnecessary package rebuild when calling make back to back. Before this change, tc-mod-iptables will be built after the main tc binary packages. Both tc-tiny and tc-full depend on tc-mod-ipables. If make is called after the packages are already built, it will check the timestamps of both packages, and will rebuild the main binaries, since the module package will be newer than the tc package. Calling BuildPackage,mod-iptables first ensures that its variant gets built before the other packages' variants. Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
* f2fstools: set each library package VARIANTEneas U de Queiroz2021-11-011-1/+3
| | | | | | | Set the different libf2fs packages's VARIANT, so that the right settings will be used by each different variant, if they are both being built. Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
* hostapd: avoid unnecessary package rebuildsEneas U de Queiroz2021-11-011-1/+3
| | | | | | | | | | | | | | | | | Package hostapd-common is a dependency of every other package defined in hostpad Makefile. It is currently built next to the bottom of that Makefile's package list. If you run make back to back, then check-compile will compare the hostapd-common timestamp to the variant being compiled, to decide if the varint needs to be rebuilt or not. Since the hostapd-conf package is built towards the end of the list, it will be newer than most of the variants, causing unnecessary package rebuilds. Move it to the top, so that its timestamp will be older than dependent packages, avoiding unnecessary rebuild of every selected variant. Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
* build: avoid rebuilds of unset VARIANT packagesEneas U de Queiroz2021-11-013-6/+10
| | | | | | | | | | | If a Makefile defines some packages with VARIANT set, and others without it, the latter will be built once for every different VARIANT set, each build trumping the previous one. Avoid rebuilds by only building unnamed variant packages when the first variant is built. Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
* build: kernel: avoid unnecessary package rebuildsEneas U de Queiroz2021-11-011-0/+4
| | | | | | | | | | | | | | | The $(LINUX_DIR)/.config timesptamp changes between runs of make target/compile and make target/install (which builds the image). Kernel-dependent packages and out of tree modules are built in between those runs, and they check the .config timestamp to decide if they need to be rebuilt. Save the target/compile .config to use its timestamp if the file does not change between runs. That way the subsequent kernel packages are not unnecessarily rebuilt when you run 'make' back to back. Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
* download: improve handling of invalid local filesEneas U de Queiroz2021-11-013-4/+17
| | | | | | | | | | | | | | | | | | | | | | | | | | | | 4e19cbc5533: [download: handle possibly invalid local tarballs] added a FORCE rule to downloaded files, so that they will be always checked by download.pl. As a side-effect, check-compile will fail, forcing unnecessary package rebuilds. The check-compile.txt log shows (for libxml2 for example): Considering target file '.../dl/libxml2-2.9.12.tar.gz'. ... prerequisite 'FORCE' of target '.../dl/libxml2-2.9.12.tar.gz' does not exist. Must remake target '.../dl/libxml2-2.9.12.tar.gz'. ... Giving up on target file '...libxml2-2.9.12/.prepared_...'. Giving up on target file '...libxml2-2.9.12/.configured_...'. Giving up on target file '...libxml2-2.9.12/.built'. Giving up on target file '...stamp/.libxml2_installed'. Giving up on target file '.compile'. Then the package is rebuilt even if it is not otherwise needed. To fix this, instead of always forcing the download target to be remade, check its hash first: if it matches, then the FORCE is not added. Signed-off-by: Eneas U de Queiroz <cotequeiroz@gmail.com>
* qosify: fix package section/categoryFelix Fietkau2021-11-011-3/+2
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* procd: update to git HEADDaniel Golle2021-11-011-3/+3
| | | | | | | 1056fc4 jail: elf: Use 64 bit variables for elf offsets c1976e5 jail: elf: Remove MIPS 64 warning Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* kernel: remove custom partition name patchDaniel Golle2021-11-011-36/+1
| | | | | | | | | Don't patch the kernel to expose the partition name in sysfs as it is already exposed via 'uevent'. All previous users have been converted to use 'uevent', so we can safely drop the custom patch. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* fstools: update to git HEADDaniel Golle2021-11-011-3/+3
| | | | | | | 19fd7fc libfstools: make sure file is closed on error d390744 libfstools: use uevent instead of relying on custom kernel patch Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* base-files: drop get_partition_by_name shell functionDaniel Golle2021-11-011-9/+0
| | | | | | | find_mmc_part provides a better alternative and all users of get_partition_by_name have been removed. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* mediatek: mt7623: make use of find_mmc_partDaniel Golle2021-11-011-5/+5
| | | | | | | Use find_mmc_part instead of previously introduced get_partition_by_name which requires a custom kernel patch. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* mediatek: mt7622: make use of find_mmc_partDaniel Golle2021-11-011-6/+6
| | | | | | | Use find_mmc_part instead of previously introduced get_partition_by_name which requires a custom kernel patch. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* uboot-envtools: mt7622: make use of find_mmc_partDaniel Golle2021-11-011-1/+1
| | | | Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* base-files: allow specifiying rootdev for find_mmc_partDaniel Golle2021-11-011-2/+8
| | | | | | | | Some devices got more than one mmc device. Allow specifying the root device as 2nd parameter of find_mmc_part so scripts can avoid matching irrelevant partitions on wrong mmc device. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* build: fix bpf toolchain dependency for qosifyFelix Fietkau2021-11-014-2/+17
| | | | | | Add hidden symbols to fix defaults with CONFIG_DEVEL unset Signed-off-by: Felix Fietkau <nbd@nbd.name>
* qosify: add package for simple qos based on ebpf+cakeFelix Fietkau2021-11-015-0/+209
| | | | | | | | | | | | | | qosify is simple daemon for setting up and managing CAKE along with a custom eBPF based classifier that sets DSCP fields of packets. It is configured via UCI and it supports the following features: - simple TCP/UDP port based mapping - IP address based mapping - priority boosting based on average packet size - bulk flow detection based on number of packets per second - dynamically add IP entries with timeout Signed-off-by: Felix Fietkau <nbd@nbd.name>
* bpf-headers: add a package with kernel headers for ebpfFelix Fietkau2021-11-013-0/+178
| | | | | | | | | | In order to genererate suitable kernel headers, a 5.10 kernel tree is prepared with a default config for mips. The arch is forced to mips in order to avoid issues with inline asm on various architectures in a way that doesn't involve relying on the host toolchain/headers. It also has the advantage of supporting both endian types Signed-off-by: Felix Fietkau <nbd@nbd.name>
* config: add choice for selecting the LLVM toolchain typeFelix Fietkau2021-11-012-7/+25
| | | | | | This is global for all BPF based packages Signed-off-by: Felix Fietkau <nbd@nbd.name>
* build: add HOST_OS_LINUX and HOST_OS_MACOS config symbolsFelix Fietkau2021-11-012-0/+9
| | | | | | This can be used to simplify host os tests in various places Signed-off-by: Felix Fietkau <nbd@nbd.name>
* dnsmasq: fix jail mount in case of ignore_hosts_dir being setDaniel Golle2021-11-011-3/+2
| | | | | | | | | Commit a2fcd3900c ("dnsmasq: improve init script") broke the existing handling for hosts_dir. Remove the redundant mount again to fix it. Reported-by: Hartmut Birr <e9hack@gmail.com> Fixes: a2fcd3900c ("dnsmasq: improve init script") Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* ubus: update to the latest versionFelix Fietkau2021-11-011-3/+3
| | | | | | | | | b743a331421d ubusd: log ACL init errors 2099bb3ad997 libubus: use list_empty/list_first_entry in ubus_process_pending_msg ef038488edc3 libubus: process pending messages in data handler if stack depth is 0 a72457b61df0 libubus: increase stack depth for processing obj msgs Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ipq40xx: add support for Netgear SRR60/SRS60 and RBR50/RBS50Davide Fioravanti2021-11-0112-5/+591
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The Netgear SRS60 and SRR60 (sold together as SRK60) are two almost identical AC3000 routers. The SRR60 has one port labeled as wan while the SRS60 not. The RBR50 and RBS50 (sold together as RBK50) have a different external shape but they have an USB 2.0 port on the back. This patch has been tested only on SRS60 and RBR50, but should work on SRR60 and RBS50. Hardware -------- SoC: Qualcomm IPQ4019 (717 MHz, 4 cores 4 threads) RAM: 512MB DDR3 FLASH: 4GB EMMC ETH: - 3x 10/100/1000 Mbps Ethernet - 1x 10/100/1000 Mbps Ethernet (WAN) WIFI: - 2.4GHz: 1x IPQ4019 (2x2:2) - 5GHz: 1x IPQ4019 (2x2:2) - 5GHz: 1x QCA9984 (4x4:4) - 6 internal antennas BTN: - 1x Reset button - 1x Sync button - 1x ON/OFF button LEDS: - 8 leds controlled by TLC59208F (they can be switched on/off independendently but the color can by changed by GPIOs) - 1x Red led (Power) - 1x Green led (Power) UART: - 115200-8-N-1 Everything works correctly. Installation ------------ These routers have a dual partition system. However this firmware works only on boot partition 1 and the OEM web interface will always flash on the partition currently not booted. The following steps will use the SRS60 firmware, but you have to chose the right firmware for your router. There are 2 ways to install Openwrt the first time: 1) Using NMRPflash 1. Download nmrpflash (https://github.com/jclehner/nmrpflash) 2. Put the openwrt-ipq40xx-generic-netgear_srs60-squashfs-factory.img file in the same folder of the nmrpflash executable 3. Connect your pc to the router using the port near the power button. 4. Run "nmrpflash -i XXX -f openwrt-ipq40xx-generic-netgear_srs60-squashfs-factory.img". Replace XXX with your network interface (can be identified by running "nmrpflash -L") 5. Power on the router and wait for the flash to complete. After about a minute the router should boot directly to Openwrt. If nothing happens try to reboot the router. If you have problems flashing try to set "10.164.183.253" as your computer IP address 2) Without NMRPflash The OEM web interface will always flash on the partition currently not booted, so to flash OpenWrt for the first time you have to switch to boot partition 2 and then flash the factory image directly from the OEM web interface. To switch on partition 2 you have to enable telnet first: 1. Go to http://192.168.1.250/debug.htm and check "Enable Telnet". 2. Connect through telent ("telnet 192.168.1.250") and login using admin/password. To read the current boot_part: artmtd -r boot_part To write the new boot_part: artmtd -w boot_part 02 Then reboot the router and then check again the current booted partition Now that you are on boot partition 2 you can flash the factory Openwrt image directly from the OEM web interface. Restore OEM Firmware -------------------- 1. Download the stock firmware from official netgear support. 2. Follow the nmrpflash procedure like above, using the official Netgear firmware (for example SRS60-V2.2.1.210.img) nmrpflash -i XXX -f SRS60-V2.2.1.210.img Notes ----- 1) You can check and edit the boot partition in the Uboot shell using the UART connection. "boot_partition_show" shows the current boot partition "boot_partition_set 1" sets the current boot partition to 1 2) Router mac addresses: LAN XX:XX:XX:XX:XX:69 WAN XX:XX:XX:XX:XX:6a WIFI 2G XX:XX:XX:XX:XX:69 WIFI 5G XX:XX:XX:XX:XX:6b WIFI 5G (2nd) XX:XX:XX:XX:XX:6c LABEL XX:XX:XX:XX:XX:69 Signed-off-by: Davide Fioravanti <pantanastyle@gmail.com> Signed-off-by: Robert Marko <robimarko@gmail.com> [added 5.10 changes for 901-arm-boot-add-dts-files.patch, moved sysupgrade mmc.sh to here and renamed it, various dtsi changes] Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
* base-files: add minimal mmc supportDavide Fioravanti2021-11-013-0/+35
| | | | | | | | | | | | | | | | Added minimal mmc support for helper functions: - find_mmc_part: Look for a given partition name. Returns the coresponding partition path - caldata_extract_mmc: Look for a given partition name and then extracts the calibration data - mmc_get_mac_binary: Returns the mac address from a given partition name and offset Signed-off-by: Davide Fioravanti <pantanastyle@gmail.com> Signed-off-by: Robert Marko <robimarko@gmail.com> [replace dd with caldata_dd, moved sysupgrade mmc to orbi] Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
* ipq40xx: enable CONFIG_CMDLINE_PARTITION and CONFIG_LEDS_TLC591XXDavide Fioravanti2021-11-012-0/+4
| | | | | | | | | | | | | | | | CONFIG_CMDLINE_PARTITION: Some devices with mmc like the Netgear Orbi Pro SRS60 or Netgear Orbi RBR50 needs to hardcode the partitions layout in the cmdline boot correctly CONFIG_LEDS_TLC591XX: This is needed for the led driver found in the Netgear Orbi Pro SRS60 Signed-off-by: Davide Fioravanti <pantanastyle@gmail.com> Signed-off-by: Robert Marko <robimarko@gmail.com> Signed-off-by: Shang Jia <jiash416@gmail.com> [added 5.10 config] Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
* realtek: backport GPIO IRQ index fixSander Vanheule2021-11-011-0/+21
| | | | | | | | Backport the patch queued upstream for 5.16. The patch differs slightly from the upstream patch due to an upstream change that added a convenience function. Signed-off-by: Sander Vanheule <sander@svanheule.net>
* kernel/modules: prevent bonding driver to create default bond0 interfaceFlorian Eckert2021-11-011-0/+1
| | | | | | | | | | | | | | | | | | | | When loading the bonding driver, bonding interface are automatically created on bonding module load. > ip a s bond0 > 14: bond0: <BROADCAST,MULTICAST,MASTER> mtu 1500 qdisc noop state DOWN > group default qlen 1000 > link/ether a6:f2:20:64:c1:b9 brd ff:ff:ff:ff:ff:ff This is not necessary in openwrt as we do not use this created interface. The netifd creates a bonding interface based on its network configuration name and configures this over the netifd bonding proto handler. In order to keep the overview of the interfaces clear, bonding interfaces should not be created automatically when loading this module, because they are not used anyway. Signed-off-by: Florian Eckert <fe@dev.tdt.de>
* elfutils: enable host buildLucian Cristian2021-11-011-0/+8
| | | | | | | | | | frr 8.0 needs host libelf dev add option for host build tested on x86, ramips, kirkwood Signed-off-by: Lucian Cristian <lucian.cristian@gmail.com> [changed commit author's email] Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
* ipq40xx: fix sleep clockPavel Kubelun2021-11-012-0/+58
| | | | | | | | It seems like sleep_clk was copied from ipq806x. Fix ipq40xx sleep_clk to the value QSDK defines. Signed-off-by: Pavel Kubelun <be.dissent@gmail.com> Signed-off-by: Christian Lamparter <chunkeey@gmail.com> [5.4+5.10]
* apm821xx: MBL: MR24: fetch IP address over dhcp by defaultChristian Lamparter2021-11-011-1/+1
| | | | | | | | This patch changes the default network configuration to fetch the IP addresses over dhcp instead of being statically assigned. Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
* ath79: fix UBNT Aircube AC gpiosNicolò Veronese2021-11-012-4/+4
| | | | | | | | | | GPIOs on the Aircube AC are wrong: - Reset GPIO moved from 17 to 12 - PoE Pass Through GPIO for Aircube AC is 3 Fixes: 491ae3357e10 ("ath79: add support for Ubiquiti airCube AC") Signed-off-by: Nicolò Veronese <nicveronese@gmail.com>
* ath79: add support for Letv LBA-047-CHShiji Yang2021-11-013-0/+178
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Specifications: SOC: QCA9531 650 MHz ROM: 16 MiB Flash (Winbond W25Q128FV) RAM: 128 MiB DDR2 (Winbond W971GG6SB) LAN: 10/100M *2 WAN: 10/100M *1 LED: BGR color *1 Mac address: label C8:0E:77:xx:xx:68 art@0x0 lan C8:0E:77:xx:xx:62 art@0x6 wan C8:0E:77:xx:xx:68 art@0x0 (same as the label) wlan C8:0E:77:xx:xx:B2 art@0x1002 (load automatically) TFTP installation: * Set local IP to 192.168.67.100 and open tftpd64, link lan port to computer. Rename "xxxx-factory.bin" to "openwrt-ar71xx-generic-ap147-16M-rootfs-squashfs.bin". * Make sure firmware file is in the tftpd's directory, push reset button and plug in, hold it for 5 seconds, and then it will download firmware from tftp server automatically. More information: * This device boot from flash@0xe80000 so we need a okli loader to deal with small kernel partition issue. In order to make full use of the storage space, connect a part of the previous kernel partition to the firmware. Stock Modify 0x000000-0x040000(u-boot) 0x000000-0x040000(u-boot) 0x040000-0x050000(u-boot-env) 0x000000-0x050000(u-boot-env) 0x050000-0xe80000(rootfs) 0x050000-0xe80000(firmware part1) 0xe80000-0xff0000(kernel) 0xe80000-0xe90000(okli-loader) 0xe90000-0xff0000(firmware part2) 0xff0000-0x1000000(art) 0xff0000-0x1000000(art) Signed-off-by: Shiji Yang <yangshiji66@qq.com>
* ath79: add support for Dongwon T&I DW02-412HJihoon Han2021-10-318-3/+315
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Dongwon T&I DW02-412H is a 2.4/5GHz band 11ac (WiFi-5) router, based on Qualcomm Atheros QCA9557. Specifications -------------- - SoC: Qualcomm Atheros QCA9557-AT4A - RAM: DDR2 128MB - Flash: SPI NOR 2MB (Winbond W25Q16DVSSIG / ESMT F25L16PA(2S)) + NAND 64/128MB - WiFi: - 2.4GHz: QCA9557 WMAC - 5GHz: QCA9882-BR4A - Ethernet: 5x 10/100/1000Mbps - Switch: QCA8337N-AL3C - USB: 1x USB 2.0 - UART: - JP2: 3.3V, TX, RX, GND (3.3V is the square pad) / 115200 8N1 Installation -------------- 1. Connect a serial interface to UART header and interrupt the autostart of kernel. 2. Transfer the factory image via TFTP and write it to the NAND flash. 3. Update U-Boot environment variable. > tftpboot 0x81000000 <your image>-factory.img > nand erase 0x1000000 > nand write 0x81000000 0x1000000 ${filesize} > setenv bootpart 2 > saveenv Revert to stock firmware -------------- 1. Revert to stock U-Boot environment variable. > setenv bootpart 1 > saveenv MAC addresses as verified by OEM firmware -------------- WAN: *:XX (label) LAN: *:XX + 1 2.4G: *:XX + 3 5G: *:XX + 4 The label MAC address was found in art 0x0. Credits -------------- Credit goes to the @manatails who first developed how to port OpenWRT to this device and had a significant impact on this patch. And thanks to @adschm and @mans0n for guiding me to revise the code in many ways. Signed-off-by: Jihoon Han <rapid_renard@renard.ga> Reviewed-by: Sungbo Eo <mans0n@gorani.run> Tested-by: Sungbo Eo <mans0n@gorani.run>
* kernel: add support for ESMT F25L16PA(2S) SPI-NORJihoon Han2021-10-313-2/+23
| | | | | | | | | This fixes support for Dongwon T&I DW02-412H which uses F25L16PA(2S) flash. Signed-off-by: Jihoon Han <rapid_renard@renard.ga> Reviewed-by: Sungbo Eo <mans0n@gorani.run> [refresh patches] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* build: fix various typosJosh Soref2021-10-312-2/+2
| | | | | | | | Fix typos in comment and user-facing help text. Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> [split out config changes, adjust commit message] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* scripts: fix various typosJosh Soref2021-10-319-14/+14
| | | | | | | | | This only affects typos in comments or user-facing output. Signed-off-by: Josh Soref <jsoref@users.noreply.github.com> [only picks changes to scripts, drop "commandline" replacement, fix case for "arbitrary", improve commit message] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ipq806x: ecw5410: use bootloader provided MAC-sRobert Marko2021-10-312-4/+2
| | | | | | | | | | | Currently, we are overriding the bootloader provided MAC-s as the ethernet aliases are reversed so MAC-s were fixed up in userspace. There is no need to do that as we can just fix the aliases instead and get rid of MAC setting via userspace helper. Fixes: 59f0a0f ("ipq806x: add Edgecore ECW5410 support") Signed-off-by: Robert Marko <robert.marko@sartura.hr>
* ipq806x: ecw5410: fix PCI1 radio caldataRobert Marko2021-10-311-2/+4
| | | | | | | | | | | | | | | | | | | | | | | | | | | ECW5410 has 2 QCA9984 cards, one per PCI controller. They are located at PCI adresses 0001:01:00.0 and 0002:01:00.0. Currently, pre-cal is not provided for 0001:01:00.0 at all,but for 0000:01:00.0 which is incorrect and causes the ath10k driver to not be able to fetch the BMI ID and use that to fetch the proper BDF but rather fail with: [ 12.029708] ath10k 5.10 driver, optimized for CT firmware, probing pci device: 0x46. [ 12.031816] ath10k_pci 0001:01:00.0: enabling device (0140 -> 0142) [ 12.037660] ath10k_pci 0001:01:00.0: pci irq msi oper_irq_mode 2 irq_mode 0 reset_mode 0 [ 13.173898] ath10k_pci 0001:01:00.0: qca9984/qca9994 hw1.0 target 0x01000000 chip_id 0x00000000 sub 168c:cafe [ 13.174015] ath10k_pci 0001:01:00.0: kconfig debug 0 debugfs 1 tracing 0 dfs 1 testmode 0 [ 13.189304] ath10k_pci 0001:01:00.0: firmware ver 10.4b-ct-9984-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 crc35 [ 15.492322] ath10k_pci 0001:01:00.0: failed to fetch board data for bus=pci,vendor=168c,device=0046,subsystem-vendor=168c,subsystem-device=cafe,variant=Edgecore-ECW541 from ath10k/QCA9984/hw1.0/board-2.bin [ 15.543883] ath10k_pci 0001:01:00.0: failed to fetch board-2.bin or board.bin from ath10k/QCA9984/hw1.0 [ 15.543920] ath10k_pci 0001:01:00.0: failed to fetch board file: -12 [ 15.552281] ath10k_pci 0001:01:00.0: could not probe fw (-12) So, provide the pre-cal for the actual PCI card and not the non-existent one. Fixes: 59f0a0f ("ipq806x: add Edgecore ECW5410 support") Signed-off-by: Robert Marko <robert.marko@sartura.hr>
* ipq806x: ecw5410: drop GPIO based MDIO1 nodeRobert Marko2021-10-311-19/+10
| | | | | | | | | | | | It looks like this is a leftover before there was a proper MDIO driver. Since both PHY-s are connected to the HW MDIO bus there is no reason for this to exist anymore, especially since it uses the same pins as the HW controller and has the pinmux for the set to "MDIO" so this worked by pure luck as GPIO MDIO would probe first and override the HW driver. Move the GMAC3 to simply use the same MDIO bus phandle. Signed-off-by: Robert Marko <robert.marko@sartura.hr>
* ath79: fix parallel image generation for Zyxel NBG6716André Valentin2021-10-311-5/+4
| | | | | | | | | This changes the image generation to use a unique directory. With parallel building it may occur that two concurrent jobs try to create an image which leds to errors. It also removes a needless subdirecory. Signed-off-by: André Valentin <avalentin@marcant.net>
* ramips: add support for HiLink HLK-7621A evaluation boardSergio Paracuellos2021-10-312-0/+119
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Specifications: - SoC: MediaTek MT7621AT - RAM: 256 MB (DDR3) - Flash: 32 MB SPI NOR 44MHz - Switch: 1 WAN, 4 LAN (Gigabit) - LEDs: 1 WAN, 4 LAN (controlled by PHY) - USB Ports: 1 x USB2, 1 x USB3 - WLAN: 1 x 2.4, 5 GHz 866Mbps (MT7612E) - Button: 1 button (reset) - UART Serial: UART1 as console : 57600 baud - Power: 12VDC, 1A Installation: Update openWRT firmware using internal GNUBEE uboot: https://github.com/gnubee-git/GnuBee-MT7621-uboot By HTTP: Initial uboot address is http://10.10.10.123, your address needs to be 10.10.10.x, and mask 255.255.255.0. By TFTP: Uboot is in client mode, the address of the firmware must be tftp://10.10.10.3/uboot.bin Recovery: Manufacturer provides MTK OpenWrt 14.07 source code, compile then flash it by uboot. HLK-7621A is a stamp hole package module for embedded development, users have to design IO boards to use it. MAC addresses: - u-boot-env contains a placeholder address: > mtd_get_mac_ascii u-boot-env ethaddr 03:17:73:ab:cd:ef - phy0 gets a valid-looking address: > cat /sys/class/ieee80211/phy0/macaddress f8:62:aa:**:**:a8 - Calibration data for &pcie2 contains a valid address, however the zeros in the right half look like it's not real: 8c:88:2b:00:00:1b - Since it's an evaluation board and there is no solid information about the MAC address assignment, the ethernet MAC address is left random. Signed-off-by: Chen Yijun <cyjason@bupt.edu.cn> [add keys and pcie nodes to properly support evaluation board] Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com> [remove ethernet address, wrap lines properly] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-26 21:37:39 +0000