aboutsummaryrefslogtreecommitdiffstats
path: root/target/linux
Commit message (Collapse)AuthorAgeFilesLines
* brcm63xx: apply updated compatibles for selecting board_infoAdrian Schmutzler2020-02-08102-224/+224
| | | | | | | | | | When changing compatibles in e4ba8c82947e ("brcm63xx: use more systematic names for image and DTS files"), I forgot to update them in kernel patches as well. So, do it now. Fixes: e4ba8c82947e ("brcm63xx: use more systematic names for image and DTS files") Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* brcm63xx: be more specific about SOC namesAdrian Schmutzler2020-02-087-6/+12
| | | | | | | This uses the specific SOC names for several devices where the Chip ID was used incorrectly before. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* brcm63xx: align DTS model with Makefile and compatibleAdrian Schmutzler2020-02-0815-18/+18
| | | | | | | | Since the model string in DTS is not required to derive board name anymore, we can now align it with the model name in Makefile and the compatible. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* brcm63xx: use compatible instead of manually setting board nameAdrian Schmutzler2020-02-087-590/+259
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This patch removes the translation of device model name into a board name in lib/brcm63xx.sh. The latter has been actually totally useless as we have the compatible which can be used instead of the board name (and actually is at other targets like ath79 or ramips). The change requires updating the base-files with the new identifiers based on compatible. With all "board names" replaced by the compatible, we do not need the old scripts to detect board name anymore and can also remove the obsolete treatment of "legacy" devices without DTS, as there are none of those left. So, this patch removes the target-specific board detection and uses the standard procedure in package/base-files/files/lib/preinit/02_sysinfo This also fixes several cases where the board name was not set or evaluated correctly: - asmax,ar1004g in 02_network - telsey,magic in 09_fix_crc/02_network - brcm,bcm96338gw in 02_network - brcm,bcm96338w in 02_network - brcm,bcm96348gw in 02_network - dynalink,rta1025w in 02_network - huawei,echolife-hg520v in 02_network - several cases in diag.sh The following orphaned identifiers are removed from board.d files: - dmv-s0 Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* brcm63xx: remove unneeded line break for DEVICE_PACKAGESAdrian Schmutzler2020-02-081-156/+76
| | | | | | | | | In brcm63xx image Makefile DEVICE_PACKAGE definitions are split into multiple lines with no apparent need. Merge them into one line to increase readability and maintainability. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* brcm63xx: use more systematic names for image and DTS filesAdrian Schmutzler2020-02-0891-310/+226
| | | | | | | | | | | | | | | | | | | | This changes brcm63xx target to achieve consistency between image names, device nodes names, compatible, DTS name and device name as much as possible. As with other targets, having consistent names is supposed to make life easier for developers and reduces the number of "variables" in the system. In particular, applying the DTS scheme will make it easier to find the correct file and architecture. DTS files are named based on the increasingly common soc-vendor-model scheme, using hyphens for separation as this seems to be the kernel way. Since the compatible is not used at the moment, I took the chance to also align them with this patch. For the SOC, the already existing CFE_CHIP_ID can be exploiting for all but two devices. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* kernel: fix typos in KernelPackage descriptionSungbo Eo2020-02-081-1/+1
| | | | | | | | | | | | Fixes: ed2839ac4190 ("kernel/modules: add kmod-pmbus-zl6100 module") Fixes: bbcb9de93547 ("Add package for gpio rotary encoder") Fixes: 76854589829a ("package/kernel: package kmod-input-matrixkmap") Fixes: 8bfef353850a ("kernel: rename kmod-switch-rtl8366_smi to kmod-switch-rtl8366-smi to avoid underscores in package names") Fixes: f03bf608b151 ("kernel: Add dummy sound driver") Fixes: dda5d9b78643 ("ramips: rename pwm kernel module") Signed-off-by: Sungbo Eo <mans0n@gorani.run>
* ramips: fix device name of netis WF-2881 to WF2881Sungbo Eo2020-02-074-11/+10
| | | | | | | | | | | The correct model name of WF-2881 is WF2881 without hyphen. The former used boardnames are not added to SUPPORTED_DEVICES, to make it explicit that the sysupgrade-tar image, which is newly added in the previous commit, should not be used to upgrade from older version. Signed-off-by: Sungbo Eo <mans0n@gorani.run> [adjust commit title] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ramips: use nand_do_upgrade for netis WF-2881Sungbo Eo2020-02-073-3/+19
| | | | | | | | | | WF-2881 sysupgrade image uses UBI rootfs, but still relies on default_do_upgrade. Because of this, config backup is not restored after sysupgrade. It can be fixed by switching to nand_do_upgrade and sysupgrade-tar image. default_do_upgrade does not handle sysupgrade-tar properly, so one should use factory image to upgrade from older version. Signed-off-by: Sungbo Eo <mans0n@gorani.run>
* ath79: add LED trigger for TL-WR902AC v1 WAN LEDLech Perczak2020-02-071-0/+1
| | | | | | | | | | | Inspired by commit c48b571ad708, add an LED trigger for the WAN LED on top of the TP-Link TL-WR902AC v1. Currently, only the LED on the port itself shows the link state, while the LED on top of the device stays dark. The WAN port of the device is a hybrid LAN/WAN one, hence why the LED at the port was labeled LAN. Signed-off-by: Lech Perczak <lech.perczak@gmail.com>
* ath79: add support for Atheros AR934x HS UARTDaniel Golle2020-02-072-0/+65
| | | | | | | | | AR934x chips also got the 'old' qca,ar9330-uart in addition to the 'new' ns16550a compatible one. Add support for UART1 clock selector as well as device-tree bindings in ar934x.dtsi to make use of that uart. Reported-by: Piotr Dymacz <pepe2k@gmail.com> Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* ath79: ar933x_uart: set UART_CS_{RX,TX}_READY_ORIDEDaniel Golle2020-02-071-0/+64
| | | | | | | | | | | | | | | | On AR934x this UART is usually not initialized by the bootloader as it is only used as a secondary serial port while the primary UART is a newly introduced NS16550-compatible. In order to make use of the ar933x-uart on AR934x without RTS/CTS hardware flow control, one needs to set the UART_CS_{RX,TX}_READY_ORIDE bits as other than on AR933x where this UART is used as primary/console, the bootloader on AR934x typically doesn't set those bits. Setting them explicitely on AR933x does not do any harm, so just set them unconditionally. Tested-by: Chuanhong Guo <gch981213@gmail.com Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* bcm53xx: build images for Luxul ABR-4500 and XBR-4500 routersDan Haab2020-02-072-0/+37
| | | | | | | | Luxul ABR-4500 and XBR-4500 devices are wired routers with 5 Ethernet ports and 1 USB 3.0 port. Flashing requires using Luxul firmware 6.4.0 or newer and uploading firmware using "Firmware Update" web UI page. Signed-off-by: Dan Haab <dan.haab@legrand.com>
* ath79: ag71xx: use netif_receive_skb_list on 4.19Chuanhong Guo2020-02-071-1/+19
| | | | | | | | This new function make batch processing of network packets possible, which slightly improves performance. Signed-off-by: Chuanhong Guo <gch981213@gmail.com> Tested-by: Rosen Penev <rosenp@gmail.com>
* kernel: remove duplicate flash chip definitionDaniel Golle2020-02-062-18/+16
| | | | | | | XTX XT25F128A shares it's chip ID with XM25QH128A which got identical features. Hence there it makes no sense to add it to the driver. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* ath79: use ar934x-spi for ar933x as wellChuanhong Guo2020-02-061-3/+2
| | | | | | | | ar933x appears to have the same spi controller as ar934x but it's not mentioned in datasheet at all. Use new spi driver instead to gain more flash operating performance. Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* ath79: add new ar934x spi driverChuanhong Guo2020-02-0611-15/+346
| | | | | | | | | | | A new shift mode was introduced since ar934x which has a way better performance than current bitbang driver and can handle higher spi clock properly. This commit adds a new driver to make use of this new feature. This new driver has chipselect properly configured and we don't need cs-gpios hack in dts anymore. Remove them. Signed-off-by: Chuanhong Guo <gch981213@gmail.com>
* ramips: add support for I-O DATA WN-AX1167GR2INAGAKI Hiroshi2020-02-054-0/+44
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I-O DATA WN-AX1167GR2 is a 2.4/5 GHz band 11ac router, based on MediaTek MT7621A. Specification: - SoC : MediaTek MT7621A - RAM : DDR3 128 MiB - Flash : NAND 128 MiB - WLAN : MediaTek MT7615D (2.4/5 GHz, 2T2R) - Ethernet : 5x 10/100/1000 Mbps - Switch : MediaTek MT7621A (MT7530) - LEDs/Input : 2x/3x (2x buttons, 1x slide-switch) - UART : through-hole on PCB - J5: Vcc, TX, RX, NC, GND - 57600 bps Flash instruction using initramfs image: 1. Boot WN-AX1167GR2 normally 2. Access to "http://192.168.0.1/" and open firmware update page ("ファームウェア") 3. Select the OpenWrt initramfs image and click update ("更新") button to perform firmware update 4. On the initramfs image, perform sysupgrade with squashfs-sysupgrade image 5. Wait ~120 seconds to complete flashing Notes: - configuration in DeviceTree of DBDC (Dual-Band-Dual-Concurrent) mode for MT7615D chip is not supported in mt76 driver - last 0x80000 (512 KiB) in NAND flash is not used on stock firmware - stock firmware requires "customized uImage header" by MSTC (MitraStar Technology Corp.), but U-Boot doesn't - uImage magic (0x0 - 0x3) : 0x434F4D42 (COMB) - header crc32 (0x4 - 0x7) : with data length and data crc32 - image name (0x20 - 0x37) : model ID and firmware versions - data length (0x38 - 0x3b): kernel + rootfs - data crc32 (0x3c - 0x3f) : kernel + rootfs Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com> Reviewed-by: Sungbo Eo <mans0n@gorani.run>
* ramips: add support for I-O DATA WN-DX1167RINAGAKI Hiroshi2020-02-056-4/+268
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | I-O DATA WN-DX1167R is a 2.4/5 GHz band 11ac rotuer, based on MediaTek MT7621A. Specification: - SoC : MediaTek MT7621A - RAM : DDR3 128 MiB - Flash : NAND 128 MiB - WLAN : MediaTek MT7615D (2.4/5 GHz, 2T2R) - Ethernet : 5x 10/100/1000 Mbps - Switch : MediaTek MT7621A (MT7530) - LEDs/Input : 2x/3x (2x buttons, 1x slide-switch) - UART : through-hole on PCB - J5: Vcc, TX, RX, NC, GND - 57600 bps Flash instruction using initramfs image: 1. Boot WN-DX1167R normally 2. Access to "http://192.168.0.1/" and open firmware update page ("ファームウェア") 3. Select the OpenWrt initramfs image and click update ("更新") button to perform firmware update 4. On the initramfs image, perform sysupgrade with squashfs-sysupgrade image 5. Wait ~120 seconds to complete flashing Notes: - configuration in DeviceTree of DBDC (Dual-Band-Dual-Concurrent) mode for MT7615D chip is not supported in mt76 driver - last 0x80000 (512 KiB) in NAND flash is not used on stock firmware - stock firmware requires "customized uImage header" by MSTC (MitraStar Technology Corp.), but U-Boot doesn't - uImage magic (0x0 - 0x3) : 0x434F4D43 (COMC) - header crc32 (0x4 - 0x7) : with data length and data crc32 - image name (0x20 - 0x37) : model ID and firmware versions - data length (0x38 - 0x3b): kernel + rootfs - data crc32 (0x3c - 0x3f) : kernel + rootfs Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com> Reviewed-by: Sungbo Eo <mans0n@gorani.run>
* ramips: extend and rename wr1201-factory-headerINAGAKI Hiroshi2020-02-051-14/+16
| | | | | | | | | | | | | | | | | | | | | This commit adds the ability to set custom uImage magic to Build/wr1201-factory-header and renames it to "Build/custom-initramfs-uimage". Custom uImage header in initramfs image is required on following devices: - I-O DATA WN-AX1167GR2 - I-O DATA WN-AX2033GR - I-O DATA WN-AX2033GR2 - I-O DATA WN-DX1167R While at it, fix typo in comment. Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com> Reviewed-by: Sungbo Eo <mans0n@gorani.run> [shorten commit title, minor commit message adjustments] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* brcm2708: update to latest patches from RPi foundationÁlvaro Fernández Rojas2020-02-04618-65/+3580
| | | | Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
* kernel: add support for GD25D05 SPI NORKoen Vandeputte2020-02-043-0/+87
| | | | | | | | | | | | | | | | | | | | | | This chip is used on newer RB912UAG-5HPnD r2 boards: Before: [ 0.642553] m25p80 spi0.0: unrecognized JEDEC id bytes: c8, 40, 10 [ 0.649381] NAND flash driver for the RouterBOARD 91x series After: [ 0.641714] m25p80 spi0.0: found gd25d05, expected m25p80 [ 0.649916] m25p80 spi0.0: gd25d05 (64 Kbytes) [ 0.655122] Creating 4 MTD partitions on "spi0.0": [ 0.660164] 0x000000000000-0x00000000c000 : "routerboot" [ 0.667782] 0x00000000c000-0x00000000d000 : "hard_config" [ 0.675073] 0x00000000d000-0x00000000e000 : "bios" [ 0.682613] 0x00000000e000-0x00000000f000 : "soft_config" [ 0.690304] NAND flash driver for the RouterBOARD 91x series Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
* kernel: bump 4.19 to 4.19.101Koen Vandeputte2020-02-0439-406/+86
| | | | | | | | | | | | | | | | | | Refreshed all patches. Fixes: - CVE-2019-14896 - CVE-2019-14897 Remove upstreamed: - 023-0007-crypto-crypto4xx-Fix-wrong-ppc4xx_trng_probe-ppc4xx_.patch - 950-0202-staging-bcm2835-camera-fix-module-autoloading.patch - 001-4.22-01-MIPS-BCM63XX-drop-unused-and-broken-DSP-platform-dev.patch Compile-tested on: cns3xxx Runtime-tested on: cns3xxx Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
* kernel: bump 4.14 to 4.14.169Koen Vandeputte2020-02-0426-325/+78
| | | | | | | | | | | | | | | | | Refreshed all patches. Fixes: - CVE-2019-14896 - CVE-2019-14897 Remove upstreamed: - 023-0007-crypto-crypto4xx-Fix-wrong-ppc4xx_trng_probe-ppc4xx_.patch - 001-4.22-01-MIPS-BCM63XX-drop-unused-and-broken-DSP-platform-dev.patch Compile-tested on: cns3xxx Runtime-tested on: cns3xxx Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
* ramips: remove unnecessary execute permission bitSungbo Eo2020-02-041-0/+0
| | | | | | | | dts file does not need to be executable. 644 is enough. Fixes: f098c612b6db ("ramips: create shared DTSI for Netgear EX2700 and WN3000RP v3") Signed-off-by: Sungbo Eo <mans0n@gorani.run>
* mediatek: remove unnecessary execute permission bitSungbo Eo2020-02-043-0/+0
| | | | | | | | kernel config and patch files do not need to be executable. 644 is enough. Fixes: 01c8f2e97cc6 ("mediatek: bump to v4.19") Signed-off-by: Sungbo Eo <mans0n@gorani.run>
* ath79: WNDR3700 v1/v2: make u-boot env partition writableMichal Cieslakiewicz2020-02-042-2/+0
| | | | | | | Remove read-only flag from U-boot environment partition for Netgear WNDR3700 v1 and v2 so u-boot-envtools can modify data there. Signed-off-by: Michal Cieslakiewicz <michal.cieslakiewicz@wp.pl>
* mediatek: remove KERNEL_PATCHVER overwrite for mt7629Adrian Schmutzler2020-02-041-1/+0
| | | | | | | | Since whole target has been bumped to kernel 4.19 in 01c8f2e97cc6 ("mediatek: bump to v4.19") we do not need the overwrite in mt7629 subtarget anymore. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* mediatek: bump to v4.19John Crispin2020-02-044-1/+562
| | | | Signed-off-by: John Crispin <john@phrozen.org>
* ath79: add support for COMFAST CF-E560ACAugust Huber2020-02-035-6/+205
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This commit adds support for the COMFAST CF-E560AC, an ap143 based in-wall access point. Specifications: - SoC: Qualcomm Atheros QCA9531 - RAM: 128 MB DDR2 (Winbond W971GG6SB-25) - Storage: 16 MB NOR (Winbond 25Q128JVSO) - WAN: 1x 10/100 PoE ethernet (48v) - LAN: 4x 10/100 ethernet - WLAN1: QCA9531 - 802.11b/g/n - 2x SKY85303-21 FEM - WLAN2: QCA9886 - 802.11ac/n/a - 2x SKY85735-11 FEM - USB: one external USB2.0 port - UART: 3.3v, 2.54mm headers already populated on board - LED: 7x external - Button: 1x external - Boot: U-Boot 1.1.4 (pepe2k/u-boot_mod) MAC addressing: - stock LAN *:40 (label) WAN *:41 5G *:42 2.4G *:4a - flash (art partition) 0x0 *:40 (label) 0x6 *:42 0x1002 *:41 0x5006 *:43 This device contains valid MAC addresses in art 0x0, 0x6, 0x1002 and 0x5006, however the vendor firmware only reads from art:0x0 for the LAN interface and then increments in 02_network. They also jump 8 addresses for the second wifi interface (2.4 GHz). This behavior has been duplicated in the DTS and ath10k hotplug to align addresses with the vendor firmware v2.6.0. Recovery instructions: This device contains built-in u-boot tftp recovery. 1. Configure PC with static IP 192.168.1.10/24 and tftp server. 2. Place desired image at /firmware_auto.bin at tftp root. 3. Connect device to PC, and power on. 4. Device will fetch flash from tftp, flash and reboot into new image. Signed-off-by: August Huber <auh@google.com> [move jtag_disable_pins, remove unnecessary statuses in DTS, remove duplicate entry in 11-ath10k-caldata, remove hub_port0 label in DTS] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ath79: WNDR3700 v2: add dash before version in device nameMichal Cieslakiewicz2020-02-034-9/+9
| | | | | | | | Adapt Netgear WNDR3700v2 device identification string to ath79 naming scheme by changing from 'wndr3700v2' to 'wndr3700-v2' (affects config, makefile, init scripts and device tree definition). Signed-off-by: Michal Cieslakiewicz <michal.cieslakiewicz@wp.pl>
* mediatek: add wmac devicetree node for MT7622John Crispin2020-02-021-0/+11
| | | | | Signed-off-by: John Crispin <john@phrozen.org> Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ath79: add common definition for Device/tplink-v2Adrian Schmutzler2020-02-022-8/+13
| | | | | | | | | | | | | | | | This adds a shared definition Device/tplink-v2 to common-tp-link.mk. Though currently only one device in ath79 uses it, putting it in the common file seems more organized. The definitions are based on the implementation in ramips target, where a lot of devices is using tplink-v2-* commands already. The '-V "ver. 2.0"' suffix for Archer D50 v1 can be removed because it's default in Build/tplink-v2-image anyway. While at it, add TPLINK_HWREVADD and TPLINK_HVERSION to DEVICE_VARS, which seems to have been overlooked when adding Archer D50 v1. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ath79: rename tplink-loader-okli to tplink-safeloader-okliAdrian Schmutzler2020-02-022-10/+10
| | | | | | | | | This renames Device/tplink-loader-okli to Device/tplink-safeloader-okli since the latter more accurately describes the combination of tplink-safeloader and loader-okli use there. The old version might be confused with other uses of the okli loader. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ath79: rename Device/tplink to Device/tplink-v1Adrian Schmutzler2020-02-021-6/+6
| | | | | | | | | As we have tplink-v2-header and tplink-v2-image recipes as well, this patch renames the Device/tplink definition to Device/tplink-v1, as it's using the tplink-v1-* commands. This should provide easier distinction in the future. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* mvebu: image: align subtargets makefile namesTomasz Maciej Nowak2020-02-014-3/+3
| | | | | | | Align subtargets makefiles names to actual subtargets. Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl> Acked-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* mvebu: image: sort devices alphabeticallyTomasz Maciej Nowak2020-02-012-114/+114
| | | | | | | | This sorts the devices in image Makefiles alphabetically. Signed-off-by: Tomasz Maciej Nowak <tomek_n@o2.pl> [fixed sorting in one case, add commit message] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ramips: add support for TP-Link RE200 v2Andreas Böhler2020-02-015-0/+180
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | TP-Link RE200 v2 is a wireless range extender with Ethernet and 2.4G and 5G WiFi with internal antennas. It's based on MediaTek MT7628AN+MT7610EN. Specifications -------------- - MediaTek MT7628AN (580 Mhz) - 64 MB of RAM - 8 MB of FLASH - 2T2R 2.4 GHz and 1T1R 5 GHz - 1x 10/100 Mbps Ethernet - UART header on PCB (57600 8n1) - 8x LED (GPIO-controlled), 2x button There are 2.4G and 5G LEDs in red and green which are controlled separately. MAC addresses ------------- The MAC address assignment matches stock firmware, i.e.: LAN : *:0D 2.4G: *:0E 5G : *:0F Installation ------------ Web Interface ------------- It is possible to upgrade to OpenWrt via the web interface. Simply flash the -factory.bin from OEM. In contrast to a stock firmware, this will not overwrite U-Boot. Serial console -------------- Opening the case is quite hard, since it is welded together. Rename the OpenWrt factory image to "test.bin", then plug in the device and quickly press "2" to enter flash mode (no line feed). Follow the prompts until OpenWrt is installed. Unfortunately, this devices does not offer a recovery mode or a tftp installation method. If the web interface upgrade fails, you have to open your device and attach serial console. Additonal notes --------------- It is possible to flash back to stock by using tplink-safeloader to create a sysupgrade image based on a stock update. After the first boot, it is necessary upgrade to another stock image, otherwise subsequent boots fail with LZMA ERROR 1 and you have to attach serial to recover the device. Signed-off-by: Andreas Böhler <dev@aboehler.at> [remove DEVICE_VARS change] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ramips: collect and harmonize TP-Link image variants in common fileAdrian Schmutzler2020-02-014-114/+100
| | | | | | | | | | | | This moves the various variants of common device definitions for TP-Link devices to a common Makefile common-tp-link.mk. This provides the opportunity to reorganize and move parameters between individual device definitions and the common ones. While at it, also use the common definitions for previously independent definitions where appropriate. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* kernel: connmark set-dscpmark follow upstreamimg attemptKevin Darbyshire-Bryant2020-01-313-159/+265
| | | | | | | | | | | | | | I'm having another attempt at trying to getting the 'store dscp into conntrack connmark' functionality into upstream kernel, since the restore function (act_ctinfo) has been accepted. The syntax has changed from 'savedscp' to 'set-dscpmark' since that conforms more closely with existing functionality. 4.14 backport is more of a hack since the structure versioning mechanism isn't in place. Signed-off-by: Kevin Darbyshire-Bryant <ldir@darbyshire-bryant.me.uk>
* ath79: add support for Ubiquiti NanoStation Loco M (XW)Roger Pueyo Centelles2020-01-314-0/+39
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This commit adds support for the NanoStation Loco M2/M5 XW devices on the ath79 target (support was long ago available on ar71xx). Specifications: - AR9342 SoC @ 535 MHz - 64 MB RAM - 8 MB SPI flash - 1x 10/100 Mbps Ethernet, 24 Vdc PoE-in - AR8032 switch - 2T2R 5 GHz radio, 22 dBm - 13 dBi built-in antenna - POWER/LAN green LEDs - 4x RSSI LEDs (red, orange, green, green) - UART (115200 8N1) on PCB Flashing via TFTP: - Use a pointy tool (e.g., pen cap, paper clip) and keep the reset button on the device or on the PoE supply pressed - Power on the device via PoE (keep reset button pressed) - Keep pressing until LEDs flash alternatively LED1+LED3 => LED2+LED4 => LED1+LED3, etc. - Release reset button - The device starts a TFTP server at 192.168.1.20 - Set a static IP on the computer (e.g., 192.168.1.21/24) - Upload via tftp the factory image: $ tftp 192.168.1.20 tftp> bin tftp> trace tftp> put openwrt-ath79-generic-xxxxx-ubnt_nanostation-loco-m-xw-squashfs-factory.bin Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
* ath79: do not set inherited phy-mode/status properties againAdrian Schmutzler2020-01-3114-22/+2
| | | | | | | | | There are several cases where phy-mode and status properties are set again in DTS(I) files although those were set to the same values in parent DTSI files already. Remove those cases (and thus also stop their proliferation by copy/paste). Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ath79: harmonize ethernet-phy naming schemeAdrian Schmutzler2020-01-316-15/+25
| | | | | | | | | | | A minority of ethernet-phy definitions seems to use numbers in label, name and reg property relatively random. This patch aligns their use to have the same numeric value for all of them. While at it, improve order of properties/add newlines for the ethX nodes where necessary. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* mediatek: mt7623: remove left-over fragments of mt7623a-rfb-emmcAdrian Schmutzler2020-01-303-468/+2
| | | | | | | | | | | | | | The image creation for the mt7623a-rfb-emmc has been removed during a patch refresh without specific comment. The corresponding base-files entries and DTS patches for 4.14 are still there. Since mt7623 is pretty dead and nobody has missed this device, let's just remove the rest. Fixes: 050da2107a7e ("mediatek: backport upstream mediatek patches") Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de> Acked-by: John Crispin <john@phrozen.org>
* ramips: use tpt DTS trigger for TP-Link TL-MR3020 v3 and TL-WA801ND v5Jan Alexander2020-01-303-7/+5
| | | | | | | | | | | | | | This converts the TP-Link TL-MR3020v3 board to use the WLAN throughput LED trigger in order to react to all VAPs. It also moves the WLAN trigger config of the TP-Link TL-WA801NDv5 to the DTS and merges the now identical LAN LED configs. Verified these changes on a TL-MR3020v3. Signed-off-by: Jan Alexander <jan@nalx.net> [changed commit title and extended commit message] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* Revert "ramips: fix NETGEAR R6260 EEPROM offset"David Bauer2020-01-304-26/+4
| | | | | | | | | | | | | This reverts commit 4716c843d62a5a4c37e7886f54bcf446aa26e786. Netgear seems to use different partition layouts on the R6260, which would require us to dynamically detect the position of (at least) the factory partition. Revert this fix to avoid breaking existing installations until a better solution has been worked out. Signed-off-by: David Bauer <mail@david-bauer.net>
* kernel: add support for XTX xt25f128 SPI-NOR flash chipsDaniel Golle2020-01-302-0/+86
| | | | | | | This fixes support for Teltonika RUT9xx which in recent versions of the device uses xt25f128b flash. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* ramips: remove duplicate DEVICE_PACKAGES for TP-Link Archer C20iAdrian Schmutzler2020-01-291-1/+0
| | | | | | | | | DEVICE_PACKAGES is specified twice for the same device. Remove the first (=older) assignment. Fixes: 40692f0fb55c ("ramips: mt7620: select only the matching mt76 driver") Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ath79: add support for GL.iNet 6408/6416 (GL.iNet V1)Adrian Schmutzler2020-01-295-0/+211
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This ports the GL.iNet 6408/6416 from ar71xx. The GL-Connect GL.iNet v1 routers are basically a TP-Link TL-WR710N with more DRAM/Flash and console/GPIO header in the same small form-factor. Specifications: - SoC: Atheros AR9331 - CPU: 400 MHz - Flash: 8/16 MiB - RAM: 64 MiB - WiFi: 2.4 GHz b/g/n (SoC) - Ethernet: 2x 100M ports (LAN/WAN) - USB: 1x 2.0 The difference between 6408 and 6416 is just the flash size. It looks like only the 16 MiB version has been advertised, while the 6408 is a modified version. There are also 1-port versions sold by third parties. Installation: Install the sysupgrade image via stock firmware GUI or upload it via uboot (web-based). The device will be available at 192.168.1.1. Attention: In ar71xx, the same board name is used for both flash versions. So, please make sure you flash the correct ath79 image when upgrading. This has been device-tested on a GL.iNet 6416. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ramips: fix NETGEAR R6260 EEPROM offsetDavid Bauer2020-01-294-4/+26
| | | | | | | | | | The EEPROM offset for the NETGEAR R6260 is incorrect, thus no valid calibration data is used. Fix this only for the NETGEAR R6260, as it's currently unknown whether or not other boards are affected. Signed-off-by: David Bauer <mail@david-bauer.net>