summaryrefslogtreecommitdiff
path: root/target/linux/ramips
Commit message (Collapse)AuthorAgeFilesLines
* ramips: fix sdhci support on mt7621John Crispin2016-12-012-24/+10
| | | | | | the IRQ level was wrong. refresh the patch while at it. Signed-off-by: John Crispin <john@phrozen.org>
* ramips: RB750Gr3: Add pwr LED and buzzer to DTSAndrew Yong2016-12-012-2/+13
| | | | | | | | | use pwr LED in diag.sh Expose unused pinmux pins as GPIOs export power LED and buzzer pins Use rb750gr3:blue:pwr LED in diag.sh for boot status instead of rb750gr3:green:usr Signed-off-by: Andrew Yong <me@ndoo.sg>
* treewide: fix button keys codes used in dtsMathias Kresin2016-11-277-7/+7
| | | | | | Use keycodes that matches the intended use case based on the label. Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: add size checks/append metadata where missingMathias Kresin2016-11-273-30/+22
| | | | Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: fix Airlink AR670W factory imageMathias Kresin2016-11-272-8/+7
| | | | | | | Move the mkwrgimg build step to the makefile. It will be required by other boards as well. Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: Add missing chunked-io directive to W2914NSV2.dtsiYong-hyu, Ban2016-11-261-0/+1
| | | | | | | | | Fix bug that LEDE failed to boot with this message: WARNING: CPU: 2 PID: 1 at drivers/spi/spi-mt7621.c:214 mt7621_spi_transfer_one_message+0x28c/0x620() Signed-off-by: Yong-hyu, Ban <perillamint@gentoo.moe>
* ramips: SamKnows SK-WB8 DTS cleanupAndrew Yong2016-11-261-11/+6
| | | | | | Use gpio.h definition of GPIO_ACTIVE_HIGH and GPIO_ACTIVE_LOW. Remove unused backup partition to increase available JFFS space. As long as U-Boot env variable "bootcount" is < 3 (reset to 0 after boot by init script) SamKnow's U-Boot will not attempt to boot from the backup flash address (0xe30000). Signed-off-by: Andrew Yong <me@ndoo.sg>
* ramips: remove kmod-ledtrig-usbdev from recently added devicesMathias Kresin2016-11-241-2/+2
| | | | | | None of the devices uses an usb led at all. Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: Add device DLINK DWR-512-BGiuseppe Lippolis2016-11-235-0/+142
| | | | | | | | | | | | | | | | | | | | | | The Dlink DWR-512-B modem is a ralink 5350 processor based embedding a 3G mini-pcie router. The oem JBOOT bootloader has to be replaced by a RT5350 SDK U-Boot such as https://github.com/stevenylai/ralink_sdk - U-Boot configured for the RT5350 256MiB SDR. Main reason to change the bootloader is the encrypted header used to store the kernel image. In this way an image can only be generated using the propietary binboy tool (included in the GPL distribution from Dlink). The binboy tool doesn't allow to modify the kernel/rootfs partition scheme. This is considered a big constraint (limited kernel size and inefficient usage of flash space). For interested people I pubblished the details of my investigation about the encrypted firmware header here: http://lists.infradead.org/pipermail/lede-dev/2016-October/003435.html Signed-off-by: Giuseppe Lippolis <giu.lippolis@gmail.com>
* ramips: do not append metadata to CY-SWR1100 factory imageMathias Kresin2016-11-231-1/+1
| | | | Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: cleanup dts files of mt7621 based boardsMathias Kresin2016-11-2320-184/+130
| | | | | | | | | | | | | | | | | | | | Fix a typo in mt7621.dtsi compatible string. Disable spi, sdhci and pci in mt7621.dtsi and enable the nodes in the indiviual board dts files. The nodes require further device specific configuration anyway. Remove the m25p80@0 spi child node from mt7621.dtsi and add the chunked-io parameter to the individual board dts files. Fix the spi flash compatible string for the WNDR3700V5. Drop the mt7621-eval-board compatible string for all boards which are not the eval board. Drop the linux,modalias parameter from spi flash node. Remove the xhci node from board files, it is already enabled in dtsi. Disable xhci for boards not having usb ports populated. Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: order mt7621.mk alphabeticalMathias Kresin2016-11-231-94/+94
| | | | Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: fix wrong check for MT7621ATAndrew Yong2016-11-211-1/+1
| | | | | | | fix in sdhci Use ralink_soc == MT762X_SOC_MT7621AT instead of CONFIG_SOC_MT7621 which is wrong and breaks builds on mt7620a-similar platforms (MT7621, MT7688) Signed-off-by: Andrew Yong <me@ndoo.sg>
* ramips: Fix sdhci kernel panics on MT7621Andrew Yong2016-11-211-2/+2
| | | | | | | | | | | | | | Enable work-arounds present in the code commented-out but needed to write to sdcard on mt7621 which currently causes kernel to oops when engaging in serious writing to sdcard. With this change applied, there are still occasional warnings thrown by the mmc driver, however, at least it no longer crashes the system and even large writes (full-card dump/erase/write/compare) don't show any corruption. Signed-off-by: Daniel Golle <daniel@makrotopia.org> Rebase to LEDE and added "CONFIG_SOC_MT7621" check to ensure non-MT7621 devices do not face performance regressions. Signed-off-by: Andrew Yong <me@ndoo.sg>
* ramips: add support for WeVO W2914NS v2perillamint2016-11-218-0/+168
| | | | Signed-off-by: Yong-hyu Ban <perillamint@gentoo.moe>
* ramips: append metadata to imagesFelix Fietkau2016-11-197-22/+33
| | | | | | | metadata checks are not enforced yet, because some images are still being built by legacy code Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ramips: add support for ZyXEL Keenetic VivaVitaly Chekryzhev2016-11-187-0/+184
| | | | | | | | | Factory image can be installed via Zyxel WebUI. Signed-off-by: Vitaly Chekryzhev <13hakta@gmail.com> [removed linux,modalias parameter from flash node in dts] [removed sdhci node from dts; no sd card slot here] Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: add MT7620 MIB support for switch and portVitaly Chekryzhev2016-11-172-72/+351
| | | | | | | | | Move definitions to header. Replace array size definitions with macro. Signed-off-by: Vitaly Chekryzhev <13hakta@gmail.com> [merged into 0513-net-mediatek-add-swconfig-driver-for-gsw_mt762x.patch] Signed-off-by: Mathias Kresin <dev@kresin.me> Acked-by: John Crispin <john@phrozen.org>
* ramips: add support for Sitecom WLR-6000Jasper Scholte2016-11-176-1/+211
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The Sitecom firmware upgrade file has SENAO_FIRMWARE_TYPE 2 set. This looks rather wrong since SENAO_FIRMWARE_TYPE 2 is kernel only but the file is way to big for only including a kernel. The factory image need to have the dlf file extension. Otherwise the Sitecom firmware rejects the file. The stock firmware uses the following mac addresses: LAN: 00:0C:F6:AA:BB:D8 (u-boot env: ethaddr) 2,4: 00:0C:F6:AA:BB:D8 (EEPROM) 5: 00:0C:F6:AA:BB:DC (EEPROM) WAN: 00:0C:F6:AA:C8:43 (u-boot env: wanaddr) Assuming the mac address range :D8 to :DC is reserved for this device, the MAC addresses were reorder to have a unique MAC address for each interface: 2.4GHz: 00:0C:F6:AA:BB:D8 LAN: 00:0C:F6:AA:BB:D9 WAN: 00:0C:F6:AA:BB:DA 5 GHz: 00:0C:F6:AA:BB:DC The first MAC is assigned to the 2.4GHz WiFi interface to keep compatibility with the SSIDs printed on the case, which have the last three sextets of the MAC address appended. There are still issues with the rt2x00 driver. It is not possible to use both wireless interfaces at the same time. The 2.4 GHz wireless (PCIe) only works if the internal 5GHz wireless is/has been enabled or used for scanning. The internal 5GHz wireless only works if the 2.4GHz wireless (PCIe) was never enabled. Disabling the 2.4Ghz after it was enabled will result in stations seeing the 5Ghz AP but are unable to connect. Due to the not optimal working wifi the manufacture, backup and storage partitions of the OEM firmware are kept for now to allow an easy switch back to the Sitecom firmware. Signed-off-by: Jasper Scholte <NightNL@outlook.com> Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: add support for Digineo AC1200 ProDominik Menke2016-11-176-4/+33
| | | | | | This is basically a ZBT-WG3526 (mt7621 SoC), but with 32M flash. Signed-off-by: Dominik Menke <dom@digineo.de>
* ramips: cleanup ZBT-WG3526.dtsiDominik Menke2016-11-171-6/+6
| | | | | | | | - replace magic numbers with GPIO_* constants - replace generic PCI definitions with correct vendor and device identification Signed-off-by: Dominik Menke <dom@digineo.de>
* ramips: split ZBT-WG3526.dts into dtsi and dts filesDominik Menke2016-11-172-112/+120
| | | | Signed-off-by: Dominik Menke <dom@digineo.de>
* ar71xx, ramips: reduce CPU load and flickering on devices using rssliedsMartin Weinelt2016-11-161-2/+2
| | | | | | | | | Polling every 40ms causes more than 10% CPU load on weak devices. An interval of 200ms is much more reasonable. Signed-off-by: Martin Weinelt <martin@darmstadt.freifunk.net> [Matthias Schiffer: adapt OpenWrt patch; add ramips; extend commit message] Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
* ramips: fix polling switch interruptsFelix Fietkau2016-11-142-2/+2
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ramips: add port index in switch config for buffalo WHR-* and WSR-* devicesFelix Fietkau2016-11-141-4/+7
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ramips: add hack to detect missing mt7621 cpu coresFelix Fietkau2016-11-141-0/+61
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ramips: fix status LED for WHR-300HP2Felix Fietkau2016-11-141-1/+1
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* treewide: dts: use keycode defines from input dt-bindingMathias Kresin2016-11-13168-337/+673
| | | | | | | | | | | | | | | | | | | | | All compiled device tree files not mentioned are binary identical to the former ones. Fix the obvious decimal/hex confusion for the power key of ramips/M2M.dts. Due to the include of the input binding header, the BTN_* node names in: - ramips/GL-MT300A.dts - ramips/GL-MT300N.dts - ramips/GL-MT750.dts - ramips/Timecloud.dts will be changed by the compiler to the numerical equivalent. Move the binding include of lantiq boards to the file where they are used the first time to hint the user where the values do come from. Signed-off-by: Mathias Kresin <dev@kresin.me>
* treewide: dts: use C style includesMathias Kresin2016-11-123-3/+3
| | | | | | | No functional change but allows me to use one script to compile all device tree source files. Signed-off-by: Mathias Kresin <dev@kresin.me>
* treewide: dts: fix dtc compiler warningsMathias Kresin2016-11-122-0/+6
| | | | | | | | | | | Fixes invalid device tree parameters. Drop the mvsw61xx node used in mvebu device tree source files. It looks like some kind of ethernet switch cargo cult. Neither the marvell,88e6352 nor the marvell,88e6172 compatible strings can be found in any LEDE file or in the kernel sources. Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: add TEW-691GR/TEW-692GR 2.4 GHz interface mac addressesMathias Kresin2016-11-092-2/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Assign an unused MAC addresses to the 2.4GHz wifi interface as it was originally planed but not possible. The MAC address numbering of the TEW-691GR changes to the following pattern: LAN: AA:BB:CC:DD:EE:0C WIFI: AA:BB:CC:DD:EE:0D WAN: AA:BB:CC:DD:EE:0F The MAC address numbering of the TEW-692GR changes to the following pattern: LAN: AA:BB:CC:DD:EE:e4 WAN: AA:BB:CC:DD:EE:e5 2.4GHz: AA:BB:CC:DD:EE:e7 5 GHz: AA:BB:CC:DD:EE:e8 Set the label and compatible string for the TEW-692GR PCIe wireless according to the the PCI binding documentation. Use the wifi@0,0 label and the pci0,0 compatible string since the PCI vendor and device id is unknown. It should work anyway since the compatible string isn't evaluated (yet). Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: use the ralink,mtd-eeprom device tree propertyMathias Kresin2016-11-097-33/+26
| | | | | | | | | | | | | | | | | | | | | Use the ralink,mtd-eeprom instead of invoking the userspace firmware loader. Set the label and compatible string according to the the PCI binding documentation. Use the wifi@0,0 label and the pci0,0 compatible string in case the PCI vendor and device id is unknown. It should work anyway since the compatible string isn't evaluated (yet). This commit might fixes the PCIe wireless for the Buffalo WHR-600D. This board was mentioned in the board 10-rt2x00-eeprom firmware hotplug script but never had the correct eeprom name set to trigger the firmware from flash extraction. Use the usual eeprom for the soc wmac of the Dovado Tiny AC. Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: Add support for Wavlink WL-WN575A3Roger Pueyo Centelles2016-11-086-0/+148
| | | | | | | | | | | | | | | | | | | | | | | | | | This commit adds support for the Wavlink WL-WN575A3, a dual-band wall-plug wireless router with the following specifications: - CPU: MediaTek MT7628AN 580MHz - Flash: 8MB - RAM: 64MB - Ethernet: 2x 10/100 Mbps (switched) - 2.4 GHz: 802.11b/g/n SoC, MIMO 2x2, 20 dBm - 5 GHz: 802.11a/n/ac MT7612E, MIMO 2x2, 20 dBm - Antennas: 4x external (2 per radio), non-detachable - LEDs: 4 programmable + LAN, WAN, POWER - Buttons: reset, WPS Flashing instructions: Factory U-boot launches a TFTP client if WPS button is pressed during power-on. Rename the sysupgrade file and configure a TFTP as follows: - Client (WL-WN575A3) IP: 192.168.10.101 - Server IP: 192.168.10.100 - Filename: firmware.bin Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
* ramips: fix Newifi D1 profileBangLang Huang2016-11-081-1/+1
| | | | | | the images were generated for the wrong flash size. Signed-off-by: BangLang Huang <banglang.huang@foxmail.com>
* ramips: Use MT7621 I2C driver for MT7628/MT7688Sven Schwermer2016-11-031-2/+4
| | | | | | | The i2c-ramips driver does not work for the MT7628/MT7688 SoCs, the i2c-mt7621 driver does. Signed-off-by: Sven Schwermer <sven.schwermer@arcor.de>
* ramips: add support for MikroTik hEX v3 (RB750Gr3)Andrew Yong2016-11-026-0/+137
| | | | | | | | | | | | | | The MikroTik hEX v3 (RB750Gr3) is a MT7621AT board which is similar to most MT7621 reference designs, it can be easily supported by this patch; however, the stock RouterBOOT bootloader has to be replaced by a MT7621 SDK U-Boot such as https://github.com/ndoo/RB750Gr3-U-Boot - U-Boot configured for the RB750Gr3 (16MiB SPI flash, 256MiB DDR3 RAM at 1200MHz). RouterBOOT, the stock bootloader, does not initialize the UART and boots silently, making it preferable to replace it with a MT7621 SDK U-Boot with UART (57600 8N1) that supports HTTP, TFTP or serial upload of sysupgrade firmware and U-Boot. Furthermore, RouterOS, the stock firmware, is contained in a proprietary modification of SquashFS without GPL sources; UART is also disabled in stock firmware. The combination of LEDE firmware generated by this PR and MT7621 SDK U-Boot expects the printed MAC address to reside at offset `0xe000` of the factory partition (absolute offset is `0x4e000`); this is similar to the factory MAC address offset for several other MT7621 devices. A 16MiB flash dump suitable for use with flashrom will be provided if/once this patch is accepted and binaries are built by LEDE buildbot. Alternatively, writing the U-Boot to the SPI flash starting at 0x0 offset and booting the board with serial console attached will allow TFTP, HTTP or serial upload of sysupgrade firmware. Signed-off-by: Andrew Yong <me@ndoo.sg>
* kernel: update kernel 4.4 to version 4.4.30Stijn Segers2016-11-021-3/+3
| | | | | | | | | | This patch bumps the 4.4 kernel from .28 to .30 and refreshes the patches. Compile-tested on ar71xx, x86/64, ramips/mt7621, brcm47xx and kirkwood. Run-tested on ar71xx & ramips/mt7621, brcm47xx and kirkwood (last two confirmed by P. Wassi). Signed-off-by: Stijn Segers <francesco.borromini@inventati.org>
* ramips: add usb packages into DIR-860L B1 profileStijn Segers2016-11-021-0/+1
| | | | | | | This patch adds the kmod-usb3 and kmod-usb-ledtrig-usbport packages to the DIR-860L B1 profile. The DIR-860L B1 has a USB 3 port. Signed-off-by: Stijn Segers <francesco.borromini@inventati.org>
* ramips: Archer C50 cleanupHenryk Heisig2016-10-313-24/+54
| | | | | | | | | | | - setting read-only flag to important partitions - enabling PA to improve 2.4 GHz signal strength - add missing leds - rename colour led - add mac adress to 5GHz wlan interface - included <dt-bindings/input/input.h> and <dt-bindings/gpio/gpio.h> Signed-off-by: Henryk Heisig hyniu@o2.pl
* ramips: Add RTC driver to kernel for working hctosysRosen Penev2016-10-312-2/+3
| | | | | | | Build the RTC driver into the kernel, (and remove the optional module), in order to make hctosys working. (Currently the module is loaded after hctosys has failed previously) Signed-off-by: Rosen Penev <rosenp@gmail.com>
* ramips: fix PSG1218 LEDsXuefu Lin2016-10-263-5/+6
| | | | Signed-off-by: Xuefu Lin <xuefulin@gmail.com>
* switch to the new usbport LED triggerRafał Miłecki2016-10-1910-36/+36
| | | | | | | This makes init.d script handle existing UCI entries using the new trigger. It also switches all targets to use its package. Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
* ramips: add support for Planex VR500.YuheiOKAWA2016-10-157-1/+112
| | | | | | | | | | | | SOC: MT7621A RAM: 256MiB NOR: MX25L51245G Non Wireless Router. Issue: soft reboot problem. SPI Flash do not exit 4byte address mode. Signed-off-by: Yuhei Okawa <tochiro.srchack@gmail.com>
* ramips: use rfkill for wps button on wlan only boardsMathias Kresin2016-10-157-29/+8
| | | | | | | | | | | | | All of the touched boards don't have an ethernet port. Allow to use the wps button on these boards to enable the wireless radio after boot. The force enabled wireless for the DCH-M225 is removed. It is reckless to bring up an unencrypted wireless network by default these days. Using the wps button to bring up the radio seam to me the better approach. Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: use rootfs splitter and new image build code for BR-6475NDMathias Kresin2016-10-152-43/+11
| | | | | | | Use the edimax uImage splitter instead of the fixed size rootfs partition. Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: move edimax images to the new build codeMathias Kresin2016-10-153-32/+37
| | | | | | | | | | | | | | | | Images installed via TFTP recovery or the Edimax webinterface of the 3g-6200n(l) are writting with the edimax header to flash. Use only one image type for these boards. The migration to the factory only images need to be done via TFTP recovery. Use the same start address for the 3g-6200n(l) factory images as the stock firmware images. Thanks to Jan Dostrasil for reporting all the issues of the 3g-6200nl and the patient testing of all changes. Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: improve edimax 6200n/nl supportMathias Kresin2016-10-154-12/+20
| | | | | | | | | | Assign the reset functionality to the wps/reset buttons. Use the wlan switch of the 6200n to enable/disable wlan. Add the internet led of the 6200nl and use the led for boot status indication Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: fix edimax 6200nl switch configMathias Kresin2016-10-152-2/+1
| | | | | | | | | | | | | With 3a9752bbd2bc00e407b5dcad21789720f40abf0e and later changes to ramips_set_preinit_iface() the default vlan config applied during preinit was changed. These changes were made without updating the default network config to ensure that vlan interfaces used for lan/wan are still configured. Fix the issue by using the default all LAN portmap and disabling not connected switch ports using portdisable device tree parameter. Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: tag the CPU switch portMathias Kresin2016-10-151-1/+1
| | | | | | | | | | | | With de5102651501e51c036f88fe9c4dbf6974e5158a the explicit tagging of the CPU port was removed. This leads to a not working default network config, with vlans enabled but disabled CPU tagging, for boards where all switch ports are having the same role. In case the ports are having different roles set, tagging is is set implicit for the CPU port by ucidef_add_switch(). Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: remove fixed lan/wan interface for switch configsMathias Kresin2016-10-151-1/+0
| | | | | | | | | | Don't set the lan/wan interface when using ucidef_add_switch. This results into a wan interface albeit all ports a annotated as lan ports. ucidef_add_switch takes care of setting the correct lan/wan interface according to the annotation of the ports. Signed-off-by: Mathias Kresin <dev@kresin.me>