aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* hostapd: remove unused variableLeon M. George2021-01-141-1/+0
| | | | | | | | | 'enc_str' was never used. Fixes: 498d84fc4e00 ("netifd: add wireless configuration support and port mac80211 to the new framework") Signed-off-by: Leon M. George <leon@georgemail.eu>
* hostapd: run as user 'network' if procd-ujail is installedDaniel Golle2021-01-144-2/+55
| | | | | | | Granting capabilities CAP_NET_ADMIN and CAP_NET_RAW allows running hostapd and wpa_supplicant without root priviledges. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* mac80211: improve error handling when adding hostapd configDaniel Golle2021-01-142-8/+8
| | | | Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* hostapd: improve error handling when adding supplicant configDaniel Golle2021-01-142-5/+5
| | | | Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* hostapd: add multicast_to_unicast and per_sta_vifEtan Kissling2021-01-142-2/+15
| | | | | | | | This allows configuration of multicast_to_unicast and per_sta_vif options. - multicast_to_unicast requests multicast-to-unicast conversion. - per_sta_vif assigns each station its own AP_VLAN interface. Signed-off-by: Etan Kissling <etan_kissling@apple.com>
* mbedtls: add config option to compile with hkdfEtan Kissling2021-01-141-1/+18
| | | | | | | This adds a config option to allow compiling with HKDF algorithm support to support applications that require this feature. Signed-off-by: Etan Kissling <etan_kissling@apple.com>
* nf-conntrack: allow querying conntrack info in nfqueueEtan Kissling2021-01-142-1/+2
| | | | | | | | | | | This allows libnetfilter_queue to access connection tracking information by requesting NFQA_CFG_F_CONNTRACK. Connection tracking information is provided in the NFQA_CT attribute. CONFIG_NETFILTER_NETLINK_GLUE_CT enables the interaction between nf_queue and nf_conntrack_netlink. Without this option, trying to access connection tracking information results in "Operation not supported". Signed-off-by: Etan Kissling <etan_kissling@apple.com>
* kernel: drop unneeded kernel version dependencyAndy Walsh2021-01-141-1/+0
| | | | | | | | | | | The current master only supports kernel 5.4, and there is no reason to remove KERNEL_IO_URING for future kernels. Drop the unneeded dependency. Signed-off-by: Andy Walsh <andy.walsh44+github@gmail.com> [improve commit title/message] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* zstd: fix package sourceDavid Bauer2021-01-141-1/+2
| | | | | | | | | | It looks like GitHub changed the URL path for release tarballs, thus the download for the zstd package was always falling back to the OpenWrt sources mirror. Fix the GitHub URL for one which works. The file hash remains unchanged. Signed-off-by: David Bauer <mail@david-bauer.net>
* uboot-envtools: add support for Aruba AP-303 and AP-365Jan Alexander2021-01-144-4/+8
| | | | | | | | | Both devices use u-boot env variables to boot OpenWrt from its flash partition. Using u-boot envtools, it is possible to change the bootcmd back to the stock firmware partition directly from OpenWrt without attaching a serial cable or even physically accessing the device. Signed-off-by: Jan Alexander <jan@nalx.net>
* ipq806x: add support for Ubiquiti UniFi AC HDJan Alexander2021-01-146-1/+326
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Hardware -------- SoC: Qualcomm IPQ8064 RAM: 512MB DDR3 Flash: 256MB NAND (Micron MT29F2G08ABBEAH4) 32MB SPI-NOR (Macronix MX25U25635F) WLAN: Qualcomm Atheros QCA9994 4T4R b/g/n Qualcomm Atheros QCA9994 4T4R a/n/ac ETH: eth0 - SECONDARY (Atheros AR8033) eth1 - MAIN (Atheros AR8033) USB: USB-C LED: Dome (white / blue) BTN: Reset Installation ------------ Copy the OpenWrt sysupgrade image to the /tmp directory of the device using scp. Default IP address is 192.168.1.20 and default username and password are "ubnt". SSH to the device and write the bootselect flag to ensure it is booting from the mtd partition the OpenWrt image will be written to. Verify the output device below matches mtd partition "bootselect" using /proc/mtd. > dd if=/dev/zero bs=1 count=1 seek=7 conv=notrunc of=/dev/mtd11 Write the OpenWrt sysupgrade image to the mtd partition labeled "kernel0". Also verify the used partition device using /proc/mtd. > dd if=/tmp/sysupgrade.bin of=/dev/mtdblock12 Reboot the device. Back to stock ------------- Use the TFTP recovery procedure with the Ubiquiti firmware image to restore the vendor firmware. Signed-off-by: Jan Alexander <jan@nalx.net>
* uboot-rockchip: update NanoPi R2S patchesDavid Bauer2021-01-143-129/+178
| | | | | | | Update the NanoPi R2S to the latest version submitted upstream. Signed-off-by: David Bauer <mail@david-bauer.net>
* uboot-rockchip: update to v2021.01Marty Jones2021-01-145-111/+127
| | | | | | | | | | | | Update the U-Boot to version v2021.01. Run-tested: FriendlyARM NanoPi R2S Radxa Rock Pi 4 Pine64 RockPro64 Signed-off-by: Marty Jones <mj8263788@gmail.com> [format commit message] Signed-off-by: David Bauer <mail@david-bauer.net>
* iw: enable HE PHY information for iw-tinyDavid Bauer2021-01-142-28/+12
| | | | | | | | | | Currently PHY information obtained from "iw phy" lacks information about a PHYs HE capabilities when using the by default installed iw-tiny. As there are already 802.11ax supported devices, enabled printing this information for the by-default installed iw variant. Signed-off-by: David Bauer <mail@david-bauer.net>
* bcm4908: backport BCM4908 integrated switch initial supportRafał Miłecki2021-01-134-0/+225
| | | | | | | Upstream driver supports bridging ports. There is no support for crossbar setup or CPU port(s) yet. Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
* bcm4908: backport PMB (Power Management Bus) driverRafał Miłecki2021-01-133-0/+502
| | | | Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
* bcm4908: add simple PCIe reset controller supportRafał Miłecki2021-01-121-0/+40
| | | | Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
* bcm4908: backport upstream DTS patchesRafał Miłecki2021-01-129-0/+501
| | | | | | | | | | 1. Netgear R8000P DTS file 2. NAND fix 3. PCIe reset block 4. Integrated switch 5. PMB block Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
* bcm4908: initial work on the Broadcom BCM4908 targetRafał Miłecki2021-01-127-0/+623
| | | | | | | | | | | | | | | | | | | | | BCM4906, BCM4908 and BCM49408 are SoCs with 64 bit ARMv8 B53 CPUs. Upstream Linux is slowly getting support for that SoCs family so it makes sense to add target for it. This prepares initial support for: 1. Asus GT-AC5300 BCM4908 based device (4 CPUs) with 1024 MiB RAM, NAND, 8 LAN ports. 2. Netgear R8000P BCM4906 based device (2 CPUs) with 512 MiB RAM, NAND, 4 LAN ports. Flashing info will come later as we learn how to generate proper images. It isn't usable yet (it only produces a bootable kernel) so "source-only" is used. Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
* kernel: enable SRv6 support by enabling lwtunnelNick Hainke2021-01-112-1/+8
| | | | | | | | | | | | | | | | | | | | | | | | Enable the ability to use segment routing based on IPv6. It allows the packet to specify a path that the packet should take through the network. Lwtunnel allow an easy encapsulation of a package. You can just install ip-full package and use it: ip -6 route add 2003::/64 dev eth0 encap seg6 mode encap \ segs 2001::1,2002::2 An IPv6 package looks like this: [IPv6 HDR][IPv6 RH][IPv6 HDR][Data...] Netifd support: https://git.openwrt.org/?p=project/netifd.git; a=commit;h=458b1a7e9473c150a40cae5d8be174f4bb03bd39 Increases imagesize by 24.125 KiB. Therefore, only enable for devices with enough flash. Signed-off-by: Nick Hainke <vincent@systemli.org>
* tools/ccache: find libzstd using rpathThomas Nixon2021-01-101-1/+4
| | | | | | | Previously, ccache would end up using the system libzstd, which is not supposed to be a build requirement. Signed-off-by: Thomas Nixon <tom@tomn.co.uk>
* rockchip: add missing Kconfig symbolsDavid Bauer2021-01-111-0/+5
| | | | | | | | When compiling with CONFIG_ALL_KMODS enabled, compilation might stall due to unset rockchip-specific config symbols. Disable these to avoid stalling this step. Signed-off-by: David Bauer <mail@david-bauer.net>
* qemu: remove obsolete packagePaul Spooren2021-01-1014-2387/+0
| | | | | | | Instead of using an ancient qemu version in-tree the building machine should just have qemu-utils installed. Signed-off-by: Paul Spooren <mail@aparcar.org>
* build/image: warn if missing qemu-img for VDI/VMDKPaul Spooren2021-01-101-2/+6
| | | | | | | | | | | | | | | Currently `qemu-img` is used to convert raw x86 images to VDI and VMDK images, used for virtual machines. Having `qemu-img` in tree requires us to maintain an ancient version of `qemu-utils`, which recently required extra work to compile with newer compiler version. This commit prints a warning message in case `qemu-img` is missing. As a next step the in-tree version of `qemu-img` can be removed. Signed-off-by: Paul Spooren <mail@aparcar.org>
* mac80211: use hostapd PID returned from config_addDaniel Golle2021-01-102-3/+2
| | | | | | | Use PID returned from config_add instead of querying procd when adding configuration to hostapd. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* hostapd: return PID on config_add callDaniel Golle2021-01-104-4/+11
| | | | | | | | | To simplify the way netifd acquires the PIDs of wpa_supplicant and hostapd let the config_add method of both of them return the PID of the called process. Use the returned PID instead of querying procd when adding wpa_supplicant configuration. Signed-off-by: Daniel Golle <daniel@makrotopia.org>
* glibc: update to latest 2.32 commit (BZ #24973)Hans Dedecker2021-01-101-2/+2
| | | | | | 0d9793e82a Fix buffer overrun in EUC-KR conversion module (bz #24973) Signed-off-by: Hans Dedecker <dedeckeh@gmail.com>
* kernel-defaults: Manage the enabling/disabling of ZSTDFlorian Fainelli2021-01-091-0/+1
| | | | | | | | Linux 5.9 introduces support for ZSTD ramdisk and initrd compression, make sure we enable/disable the relevant options when building an initramfs enabled kernel. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
* kernel-defaults: Delete external source tree user_headersFlorian Fainelli2021-01-091-0/+1
| | | | | | | | | | When we use an external kernel tree which may not have been fully cleaned, there may be user_headers left which do not match the target architecture, leading to build failures for packages that do an explicit inclusion of user_headers (such as iproute2 or iptables). Make sure we delete them while preparing the directory. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com>
* iptables: update to 1.8.6Curtis Deptuck2021-01-097-39/+32
| | | | | | | | | | | | | Update iptables to 1.8.6 ChangeLog: https://netfilter.org/projects/iptables/files/changes-iptables-1.8.6.txt Refresh patch: 101-remove-check-already.patch Signed-off-by: Curtis Deptuck <curtdept@me.com> Signed-off-by: Hans Dedecker <dedeckeh@gmail.com> [refresh patches]
* kernel: bump 5.4 to 5.4.87John Audia2021-01-087-14/+14
| | | | | | | | | | | | | | | | | | All modification by update_kernel.sh. Build system: x86_64 Build-tested: ipq806x/R7800, bcm27xx/bcm2711 Run-tested: ipq806x/R7800 No dmesg regressions, everything functional Compile-tested [*]: ath79/{generic,tiny}, ipq40xx, octeon, ramips/mt7621, realtek, x86/64. Run-tested [*]: ramips/mt7621 (DIR-878 A1, R6800, RT-AC57U), octeon (EdgeRouter Lite). Signed-off-by: John Audia <graysky@archlinux.us> Tested-by: Stijn Segers <foss@volatilesystems.org> [*]
* realtek: add support for ZyXEL GS1900-8HP v1 and v2Stijn Segers2021-01-084-0/+40
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The ZyXEL GS1900-8HP is an 8 port gigabit switch with PoE+ support. There are two versions on the market (v1 & v2) which share similar specs (same flash size and flash layout, same RAM size, same PoE+ power envelope) but have a different case and board layout that they each share with other GS1900 siblings. The v1 seems to share its PCB and case with non-PoE GS1900-8; as such, adding support for the GS1900-8 would probably be trivial. The v2 seems to share its casing and platform with its already supported bigger brother, the GS1900-10HP - its board looks the same, except for two holes where the GS1900-10 has its SFP ports. Like their 10 port sibling, both devices have a dual firmware layout. Both GS1900-8HP boards have the same 70W PoE+ power budget. In order to manipulate the PoE+, one needs the rtl83xx-poe package [1]. After careful consideration it was decided to go with separate images for each version. Specifications (v1) ------------------- * SoC: Realtek RTL8380M 500 MHz MIPS 4KEc * Flash: Macronix MX25L12835F 16 MiB * RAM: Nanya NT5TU128M8HE-AC 128 MiB DDR2 SDRAM * Ethernet: 8x 10/100/1000 Mbit * PoE+: Broadcom BCM59111KMLG (IEEE 802.3at-2009 compliant, 2x) * UART: 1 serial header with populated standard pin connector on the left side of the PCB, towards the bottom. Pins are labeled: + VCC (3.3V) + TX + RX + GND Specifications (v2) ------------------- * SoC: Realtek RTL8380M 500 MHz MIPS 4KEc * Flash: Macronix MX25L12835F 16 MiB * RAM: Samsung K4B1G0846G 128 MiB DDR3 SDRAM * Ethernet: 8x 10/100/1000 Mbit * PoE+: Broadcom BCM59121B0KMLG (IEEE 802.3at-2009 compliant) * UART: 1 angled serial header with populated standard pin connector accessible from outside through the ventilation slits on the side. Pins from top to bottom are clearly marked on the PCB: + VCC (3.3V) + TX + RX + GND Serial connection parameters for both devices: 115200 8N1. Installation ------------ Instructions are identical to those for the GS1900-10HP and apply both to the GS1900-8HP v1 and v2 as well. * Configure your client with a static 192.168.1.x IP (e.g. 192.168.1.10). * Set up a TFTP server on your client and make it serve the initramfs image. * Connect serial, power up the switch, interrupt U-boot by hitting the space bar, and enable the network: > rtk network on * Since the GS1900-10HP is a dual-partition device, you want to keep the OEM firmware on the backup partition for the time being. OpenWrt can only boot off the first partition anyway (hardcoded in the DTS). To make sure we are manipulating the first partition, issue the following commands: > setsys bootpartition 0 > savesys * Download the image onto the device and boot from it: > tftpboot 0x84f00000 192.168.1.10:openwrt-realtek-generic-zyxel_gs1900-8hp-v{1,2}-initramfs-kernel.bin > bootm * Once OpenWrt has booted, scp the sysupgrade image to /tmp and flash it: > sysupgrade /tmp//tmp/openwrt-realtek-generic-zyxel_gs1900-8hp-v{1,2}-squashfs-sysupgrade.bin Signed-off-by: Stijn Segers <foss@volatilesystems.org> [merge PoE case, keep device definitions separate, change all those hashes in the commit message to something else so they don't get removed when changing the commit ...] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* realtek: remove model prefix from LED label for ZyXEL GS1900Adrian Schmutzler2021-01-081-1/+1
| | | | | | This is used as fixed status LED, so no migration is needed. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* realtek: introduce shared DTSI for GS1900 seriesStijn Segers2021-01-082-208/+153
| | | | | | | | | | | The ZyXEL GS1900-8HP v1, v2 and GS1900-10HP are all built on a similar Realtek RTL8380M platform. Create a common DTSI in preparation for GS1900-8HP support, and switch to the macros defined in rtl838x.dtsi. Signed-off-by: Stijn Segers <foss@volatilesystems.org> [drop redundant includes, use &mdio directly, do not replace SFP ports] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* kernel: add hid-cp2112 driver supportPawel Dembicki2021-01-071-0/+17
| | | | | | | | This patch adds kernel module for Silicon Labs CP2112 HID USB to SMBus Master Bridge. This is a HID device driver which registers as an i2c adapter and gpiochip to expose these functions of the CP2112. Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
* ramips: remove trailing whitespace in MakefilesLeon M. George2021-01-072-2/+2
| | | | | | | | Remove trailing whitespaces in two *.mk files. Signed-off-by: Leon M. George <leon@georgemail.eu> [fix title, add message] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* realtek: ZyXEL: spell as done by manufacturerStijn Segers2021-01-072-2/+2
| | | | | | | ZyXEL spells its own name all uppercase with just the Y lowercase. Adapt the realtek target to follow this (other OpenWrt targets already do so). Signed-off-by: Stijn Segers <foss@volatilesystems.org>
* realtek: move memory node to device DTSStijn Segers2021-01-076-5/+25
| | | | | | | | | Move the memory out of the rtl838x.dtsi and into the device family DTSI or device DTS if applicable. This aligns with upstream practice. Signed-off-by: Stijn Segers <foss@volatilesystems.org> [add missing block for dgs-1210-10p, move block below chosen node] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* realtek: remove duplicate '/dts-v1/;' identifierAdrian Schmutzler2021-01-072-2/+0
| | | | | | | The identifier is already present in rtl838x.dtsi, and adding it twice is not only redundant but actually wrong. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* realtek: set PoE power budget for ZyXEL GS1900-10HPStijn Segers2021-01-071-0/+3
| | | | | | | As per the manufacturer's specifications, set the GS1900-10HP PoE power budget to 77W. Signed-off-by: Stijn Segers <foss@volatilesystems.org>
* base-files: read all 3 bytes in get_magic_vfat() at onceAdrian Schmutzler2021-01-072-2/+2
| | | | | | | While the speed improvement might be negligible, there is still no reason to read individual bytes. Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ath79: make Engenius fakeroot partitions read-onlyMichael Pratt2021-01-072-0/+2
| | | | | | | | | | | | | | | | | | For: - ENH202 v1 - ENS202EXT v1 These boards were committed before it was discovered that for all Engenius boards with a "failsafe" image, forcing the failsafe image to load next boot can be achieved by editing the u-boot environment like: `fw_setenv rootfs_checksum 0` So it's not necessary to delete a partition to boot to failsafe image. Signed-off-by: Michael Pratt <mcpratt@pm.me>
* ath79: move small-flash Engenius boards to tinyMichael Pratt2021-01-079-100/+104
| | | | | | | | | | | | | | | | | | | | | | | | | This moves some of the Engenius boards from generic to tiny: - EAP350 v1 - ECB350 v1 - ENH202 v1 For these, factory.bin builds are already failing on master branch because of the unique situation for these boards: - 8 MB flash - an extra "failsafe" image for recovery - TFTP does not work (barely possible with 600 MTU) - bootloader loads image from a longer flash offset - 1 eraseblock each needed for OKLI kernel loader and fake rootfs - using mtd-concat to make use of remaining space... The manual alternative would be removing the failsafe partition. However this comes with the risk of extremely difficult recovery if a flash ever fails because TFTP on the bootloader is bugged. Signed-off-by: Michael Pratt <mcpratt@pm.me> [improve commit message] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* imx6: disable unrequired pcie host driverKoen Vandeputte2021-01-071-2/+0
| | | | | | | | | | | | | imx6 has it's own pcie host driver so we do not need the one from DW. This fixes following boot error: [ 0.156913] dw-pcie 1ffc000.pcie: IRQ index 1 not found Fixes: 6d5291ff7244 ("imx6: add support for kernel 5.4") Signed-off-by: Tim Harvey <tharvey@gateworks.com> Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
* imx6: refresh kernel configKoen Vandeputte2021-01-071-73/+2
| | | | Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
* imx6: gw52xx: fix duplicate regulator namingKoen Vandeputte2021-01-071-0/+37
| | | | | | | | | | | 2 regulator descriptions carry identical naming. This leads to following boot warning: [ 0.173138] debugfs: Directory 'vdd1p8' with parent 'regulator' already present! Fix this by renaming the one used for audio. Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
* tools/cmake: always use non-ccache CC and CXX variablesSven Wegener2021-01-061-14/+2
| | | | | | | | | | | | | | | cmake is a dependency of ccache, which means it is build before ccache is available and hence must be build with non-ccache CC and CXX. It currently works, because the cmake build system splits the compiler variable and treats them as multiple compilers to check. For "ccache gcc" it first tests for "ccache", which always fails, because ccache is not a compiler by itself, even if it is available, and then ends up calling "gcc" alone, effectively never using ccache. Let's make this explicit by forcing the use of non-ccache CC and CXX. Signed-off-by: Sven Wegener <sven.wegener@stealer.net>
* lua: add -ldl for glibc buildsRosen Penev2021-01-061-2/+2
| | | | | | | | | | | | | For glibc, lua needs an explicit link to libdl as glibc has it separate Fixes the following error in at least collectd: ld: usr/lib/liblua.so: undefined reference to `dlopen' ld: usr/lib/liblua.so: undefined reference to `dlclose' ld: usr/lib/liblua.so: undefined reference to `dlerror' ld: usr/lib/liblua.so: undefined reference to `dlsym' Signed-off-by: Rosen Penev <rosenp@gmail.com>
* build: use ccache -C for cleaning the cacheSven Wegener2021-01-062-3/+3
| | | | | | | | | | This keeps the configuration, like the size of the cache, and the statistics intact. Move the removal of the cache directory to the distclean target, but only delete the .ccache directory inside of our build tree, as we should not mess with a user-configured external ccache directory this destructively. Signed-off-by: Sven Wegener <sven.wegener@stealer.net>
* rpcd: update to git HEADDaniel Golle2021-01-071-4/+4
| | | | | | | | | | fd017ba iwinfo: add ht and vht operation info to wifi scan 4c66b31 iwinfo: export center channel for info ubus call e28d4a5 iwinfo: add support for 802.11ad and GCMP 5c15f57 iwinfo: return hwmode 'ad' on 802.11ad-only hardware ea7f471 iwinfo: include ht_operation data only if available Signed-off-by: Daniel Golle <daniel@makrotopia.org>
generated by cgit v1.2.3 (git 2.25.1) at 2024-07-05 05:17:56 +0000