aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* ramips: mt7621: fix usb initRoman Yeryomin2018-02-221-0/+1
| | | | | | | | | | In 4.14 reg-names is required to identify resources. Otherwise init fails with: [ 5.900501] xhci-mtk 1e1c0000.xhci: invalid resource [ 5.910499] xhci-mtk: probe of 1e1c0000.xhci failed with error -22 Signed-off-by: Roman Yeryomin <roman@advem.lv>
* ath9k: merge a RCU fix for station tx cleanupFelix Fietkau2018-02-221-0/+43
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* octeontx: remove lots of bogus kernel config overridesFelix Fietkau2018-02-221-73/+3
| | | | | | Some of them break when building with all modules enabled Signed-off-by: Felix Fietkau <nbd@nbd.name>
* kernel: add missing symbolTim Harvey2018-02-221-0/+1
| | | | Signed-off-by: Tim Harvey <tharvey@gateworks.com>
* octeontx: remove undefs of CONFIG_NET_VENDOR_*Tim Harvey2018-02-221-52/+0
| | | | Signed-off-by: Tim Harvey <tharvey@gateworks.com>
* octeontx: remove unnecessary CONFIG_DEBUG_INFOTim Harvey2018-02-221-1/+0
| | | | | Cc: Hauke Mehrtens <hauke@hauke-m.de> Signed-off-by: Tim Harvey <tharvey@gateworks.com>
* octeontx: add fpu supportTim Harvey2018-02-221-1/+1
| | | | | Cc: Hauke Mehrtens <hauke@hauke-m.de> Signed-off-by: Tim Harvey <tharvey@gateworks.com>
* octeontx: remove CFLAGSTim Harvey2018-02-221-1/+0
| | | | | | | | You should not define CFLAGS for the toolchain as this will also leak into other targets if they share the same toolchain. Cc: Hauke Mehrtens <hauke@hauke-m.de> Signed-off-by: Tim Harvey <tharvey@gateworks.com>
* octeontx: add FAT FS support to kernelTim Harvey2018-02-221-0/+2
| | | | | | | The CN80XX Boot firmware uses an embedded FAT12 filesystem. For some reason busybox can't mount this unless its enabled static in the kernel. Signed-off-by: Tim Harvey <tharvey@gateworks.com>
* octeontx: add USB_PCI supportTim Harvey2018-02-221-3/+4
| | | | | | The CN80XX XHCI is supported through PCI Signed-off-by: Tim Harvey <tharvey@gateworks.com>
* imx6: disable dma on uartKoen Vandeputte2018-02-221-0/+23
| | | | | | | | | | | | | | | | | | | | | | | When this target got updated to 4.14, this patch got removed to re-evaluate if it was still needed. Extensive testing now shows this issue is still present. Let's re-add the patch to fix it for now. As the uart bus is very low bandwidth .. performance impact is negligible. Boot log: [ 22.513051] imx-uart 2020000.serial: DMA transaction error. [ 22.522721] imx-uart 2020000.serial: DMA transaction error. As a sidenote: The patch mentiones an issue with RS485, but the bootlog errors above were recorded with the uart ports in standard RS232 mode. Compile/Run-tested on imx6/GW5200 Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
* kernel: refresh patchesKoen Vandeputte2018-02-228-32/+34
| | | | | | Some fuzz was introduced due to the netfilter-offload series Signed-off-by: Koen Vandeputte <koen.vandeputte@ncentric.com>
* kernel: remove kmod-appletalkFelix Fietkau2018-02-221-22/+0
| | | | | | | This has been obsolete for many years now and has been implicated in a recent build failure Signed-off-by: Felix Fietkau <nbd@nbd.name>
* kernel: fix kernel module packaging errors on imx6Felix Fietkau2018-02-221-3/+2
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* grub2: disable building platform code for target utilityFelix Fietkau2018-02-221-1/+2
| | | | | | It is not used and it was causing a build error with GCC 7.3 Signed-off-by: Felix Fietkau <nbd@nbd.name>
* toolchain/gcc: use version 7.x for x86 by defaultFelix Fietkau2018-02-212-1/+7
| | | | | | | This is needed for retpoline support, which mitigates the spectre attacks. Signed-off-by: Felix Fietkau <nbd@nbd.name>
* toolchain/gcc: remove unused config symbolFelix Fietkau2018-02-211-4/+0
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* kernel: avoid stripping retpoline flag with CONFIG_MODULE_STRIPPEDFelix Fietkau2018-02-211-11/+4
| | | | | | Fixes kernel warning when loading modules Signed-off-by: Felix Fietkau <nbd@nbd.name>
* netfilter: add a xt_FLOWOFFLOAD target for NAT/routing offload supportFelix Fietkau2018-02-214-1/+479
| | | | | | | | | | This makes it possible to add an iptables rule that offloads routing/NAT packet processing to a software fast path. This fast path is much quicker than running packets through the regular tables/chains. Requires Linux 4.14 Signed-off-by: Felix Fietkau <nbd@nbd.name>
* kernel: backport netfilter NAT offload support to 4.14Felix Fietkau2018-02-2143-0/+11343
| | | | | | | | | This only works with nftables for now, iptables support will be added later. Includes a number of related upstream nftables improvements to simplify backporting follow-up changes Signed-off-by: John Crispin <john@phrozen.org> Signed-off-by: Felix Fietkau <nbd@nbd.name>
* kernel: backport a series of netfilter cleanup patches to 4.14Felix Fietkau2018-02-2116-0/+3232
| | | | | | Preparation for backporting upstream NAT offload support Signed-off-by: Felix Fietkau <nbd@nbd.name>
* nftables: update to 0.8.2, backport flowtable supportFelix Fietkau2018-02-216-0/+1581
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* libnftnl: backport flowtable supportFelix Fietkau2018-02-213-1/+1704
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* mac80211: add minstrel improvements/fixesFelix Fietkau2018-02-214-0/+1061
| | | | | | | | - Simplify debugfs code - Reduce size - Fix handling of CCK rates Signed-off-by: Felix Fietkau <nbd@nbd.name>
* hostapd: add support for sending 802.11v disassoc imminent notifications to ↵Felix Fietkau2018-02-211-0/+97
| | | | | | clients via ubus Signed-off-by: Felix Fietkau <nbd@nbd.name>
* hostapd: add support for issuing 802.11k beacon measurement requests via ubusFelix Fietkau2018-02-211-0/+90
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* hostapd: expose client 802.11k capabilities via ubusFelix Fietkau2018-02-211-0/+6
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* hostapd: return with 80211 codes in handle event functionNick Hainke2018-02-212-23/+36
| | | | | | | | | | | If the auth or assoc request was denied the reason was always WLAN_STATUS_UNSPECIFIED_FAILURE. That's why for example the wpa supplicant was always trying to reconnect to the AP. Now it's possible to give reasoncodes why the auth or assoc was denied. Signed-off-by: Nick Hainke <vincent@systemli.org>
* hostapd: add IEEE 802.11v supportLorenzo Santina2018-02-213-2/+21
| | | | | | | | | | | | | | | | | | Add Wireless Network Management (IEEE 802.11v) support to: - hostapd-full - wpa_supplicant-full It must be enabled at runtime via UCI with: - option ieee80211v '1' Add UCI support for: - time_advertisement - time_zone - wnm_sleep_mode - bss_transition Signed-off-by: Lorenzo Santina <lorenzo.santina@edu.unito.it>
* hostapd: add support for turning on 802.11k/v features via ubusFelix Fietkau2018-02-212-7/+125
| | | | | | | Neighbor reports are enabled implicitly on use, beacon reports and BSS transition management need to be enabled explicitly Signed-off-by: Felix Fietkau <nbd@nbd.name>
* mac80211: round up tx status headroomFelix Fietkau2018-02-211-0/+26
| | | | | | Fixes unaligned access exceptions in mt76 when transmitting beacons Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ramips: fix MT7621 switch driver IRQ storm on init with linux 4.14Felix Fietkau2018-02-211-4/+7
| | | | | | | | The hardware emits some interrupts while initializing and handling them can mess up the state or cause infinite loops. Fix this by disabling IRQs during init and re-enabling them afterwards Signed-off-by: Felix Fietkau <nbd@nbd.name>
* kernel: fix MIPS GIC clock source counter width (affects MT7621)Felix Fietkau2018-02-211-0/+23
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ramips: fix MT7621 PCI driver on 4.14Felix Fietkau2018-02-211-1/+1
| | | | | | The .map_irq hook can now be called after init. Signed-off-by: Felix Fietkau <nbd@nbd.name>
* kernel: fix return value of usb_xhci_needs_pci_resetFelix Fietkau2018-02-201-2/+5
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* kernel: fix disabling common usb quirks on 4.14Felix Fietkau2018-02-201-7/+5
| | | | Signed-off-by: Felix Fietkau <nbd@nbd.name>
* ubox: fix PKG_MIRROR_HASHJohn Crispin2018-02-201-1/+1
| | | | Signed-off-by: John Crispin <john@phrozen.org>
* ramips: add support for Tama W06INAGAKI Hiroshi2018-02-205-0/+129
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Tama Electric Axing W06 is a 2.4 GHz band 11n router, based on Mediatek MT7688AN. Specification: - MT7688AN (575 MHz) - 64 MB of RAM (DDR2 SDRAM) - 16 MB of Flash (SPI) - 1T1R 2.4 GHz - 1x 10/100/1000 Mbps Ethernet - 4x LEDs (GPIO connected: 3), 1x button - 1x USB 2.0 Type-A (host) - UART header on PCB (GND, RX, TX, Vcc from RJ45 side) Flash instruction using sysupgrade image: 1. Connect micro-USB cable for power supply into W06 and turn on the router 2. Connect to wifi with SSID "tama-*" with password. Complete SSID and password are listed on the back of the router 3. Access to 192.168.1.1 and login with user name "admin" and password empty 4. In firmware update(ファームウェア更新) page, click "参照" button and click "ブラウザー" button to open file browser, select the sysupgrade image and press OK button 5. Wait ~150 seconds to complete flashing Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com>
* ramips: improve GnuBee Personal Cloud Two supportMathias Kresin2018-02-207-36/+22
| | | | | | | | | | | | | | | | | | Use the generic board detection for the GnuBee Personal Cloud Two instead of the target specific one as all recent additions are doing. Fixup the pinmux to set all pins used as GPIO to the function GPIO. Request pins where used. Drop the i2c from the dts. There is nothing connected. While at it fix an indentation issue and use references instead of duplicating the whole node path. Use the same switch config as for the GB-PC1 and drop the led trigger for the not supported IP1001 phy connected to second rgmii. Fixes: c60a21532bc9 ("ramips: Add support for the GnuBee Personal Cloud Two") Signed-off-by: Mathias Kresin <dev@kresin.me>
* ramips: use generic board detection for D-Link DAP-1522 A1Mathias Kresin2018-02-205-8/+5
| | | | | | | Use the generic board detection for the D-Link DAP-1522 A1 instead of the target specific one as all recent additions are doing. Signed-off-by: Mathias Kresin <dev@kresin.me>
* lantiq: kernel 4.14: update dts filesHauke Mehrtens2018-02-2038-159/+794
| | | | | | | | | | | | | | | | | | | | Updated the devicetree source files to make use of the following upstreamed drivers: - xrx200 ethernet phy - reset controller unit - dwc2 - fpi Use our custom xrx200 ethernet phy compatible to support boards, which have switched the vr9 revision during lifetime, with a single devicetree source file. By switching to the dwc2 driver + usb phy framework, we don't need to used our custom gpio power patch and can use a fixed regulator instead. Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de> Signed-off-by: Mathias Kresin <dev@kresin.me>
* lantiq: kernel 4.14: cleanup dts filesMathias Kresin2018-02-2055-4194/+4033
| | | | | | | Move common used parts to dtsi. Add labels to nodes and reference nodes via labels in dts files. Signed-off-by: Mathias Kresin <dev@kresin.me>
* lantiq: kernel 4.14: fix usbMathias Kresin2018-02-202-0/+62
| | | | | | | | | | On danube the USB0 registers are at 1e101000 similar to all other lantiq SoCs. On Danube and AR9 the USB core is connected to the AHB bus, hence we need to enable the AHB Bus as well. Signed-off-by: Mathias Kresin <dev@kresin.me>
* lantiq: kernel 4.14: autoselect matching vr9 rev firmwareMathias Kresin2018-02-201-0/+45
| | | | | | | | | | | | Add a custom xrx200 ethernet phy compatible to load the firmware matching the vr9 revision without specifing an expected revision. We have quite a few boards in the tree were later produced ones are using a more recent vr9. It is impossible to distinguish which revision of the vr9 is used without opening the case and removing a heatsink for some of them. Signed-off-by: Mathias Kresin <dev@kresin.me>
* lantiq: kernel 4.14: don't use CPU interrupt controller IPI IRQ domain supportMathias Kresin2018-02-201-0/+271
| | | | | | | | | | | | | | | | This reverts kernel commit 1eed40043579 ("MIPS: smp-mt: Use CPU interrupt controller IPI IRQ domain support"). With the patch applied, the kernel hangs during boot if SMP is active. The Lantiq IRQ controller gets registered first and it directly handles the MIPS native SW1/2 and HW0 - HW5 IRQs. It looks like this controller already registers IRQ 0 - 7 and the generic driver only gets the following IRQs starting later. The upstream discussion can be found at https://www.linux-mips.org/archives/linux-mips/2017-05/msg00059.html. Signed-off-by: Mathias Kresin <dev@kresin.me>
* lantiq: kernel 4.14: select MFD SYSCON for aseMathias Kresin2018-02-201-0/+24
| | | | | | Enable syscon to use it for the RCU MFD on Amazon SE as well. Signed-off-by: Mathias Kresin <dev@kresin.me>
* lantiq: kernel 4.14: update patches and configHauke Mehrtens2018-02-2035-3208/+333
| | | | | Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de> Signed-off-by: Mathias Kresin <dev@kresin.me>
* lantiq: kernel 4.14: copy patches, config and dts filesHauke Mehrtens2018-02-20114-0/+28761
| | | | | | | This just copies the patches, configuration and dts files into the directories hich are used for kernel 4.14. Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
* lantiq: fix lantiq applications kernel 4.14 compatiblityHauke Mehrtens2018-02-2013-15/+129
| | | | | | | | This is fixing multiple compile problems with kernel 4.14 and updates the code to take care of changes introduced between kernel 4.9 and 4.14. Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de> Signed-off-by: Mathias Kresin <dev@kresin.me>
* lantiq: make dts files and kernel config kernel version specificMathias Kresin2018-02-2080-2/+1
| | | | | | | | | | Move the devicetree source files to a kernel specific directory in preparation of adding kernel 4.14 support. Rename the subtarget kernel config files to match a specific kernel version. Signed-off-by: Mathias Kresin <dev@kresin.me>