aboutsummaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* toolchain: remove uClibc-ngRosen Penev2020-12-2230-593/+0
| | | | | | | | After musl was introduced, it was desired to remove uClibc-ng. As ARC has no musl support, it was kept around. However, glibc 2.32 includes ARC support. This makes it possible to finally remove it. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* toolchain: remove uClibc-ng config stuffRosen Penev2020-12-222-22/+0
| | | | | | This is in preparation for removing it. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* toolchain: remove uClibc install stuffRosen Penev2020-12-221-28/+3
| | | | | | This is preparation for removing uClibc-ng. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* toolchain: default to glibc for ARCRosen Penev2020-12-221-2/+1
| | | | | | | | glibc 2.32 gained support for the ARC architecture. This is preparation for removing uClibc-ng. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* realtek: use kernel defined haltSander Vanheule2020-12-221-7/+0
| | | | | | | | | If _machine_hang is not defined on MIPS, the kernel will check if the CPU can enter a more power efficient sleep mode. Since the realtek platform supports mips32_r2, this should issue a WAIT instruction instead of a trivial infinite loop. Signed-off-by: Sander Vanheule <sander@svanheule.net>
* hostapd: Use EAPOLv1 (802.1X-2001) if WPA enabledNick Lowe2020-12-221-1/+1
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently, EAPOLv2 (802.1X-2004) is used by default for legacy clients that are not WPA2 (RSN) capable. These legacy clients are often intolerant to this EAPOL version and fail to connect. hostapd.conf upstream documents for eapol_version the following and that this is a known compatibility issue with version 2: // IEEE 802.1X/EAPOL version // hostapd is implemented based on IEEE Std 802.1X-2004 which defines EAPOL // version 2. However, there are many client implementations that do not handle // the new version number correctly (they seem to drop the frames completely). // In order to make hostapd interoperate with these clients, the version number // can be set to the older version (1) with this configuration value. // Note: When using MACsec, eapol_version shall be set to 3, which is // defined in IEEE Std 802.1X-2010. //eapol_version=2 For the wpa parameter, hostapd.conf upstream documents that this is a bitfield, configured as follows: // Enable WPA. Setting this variable configures the AP to require WPA (either // WPA-PSK or WPA-RADIUS/EAP based on other configuration). For WPA-PSK, either // wpa_psk or wpa_passphrase must be set and wpa_key_mgmt must include WPA-PSK. // Instead of wpa_psk / wpa_passphrase, wpa_psk_radius might suffice. // For WPA-RADIUS/EAP, ieee8021x must be set (but without dynamic WEP keys), // RADIUS authentication server must be configured, and WPA-EAP must be included // in wpa_key_mgmt. // This field is a bit field that can be used to enable WPA (IEEE 802.11i/D3.0) // and/or WPA2 (full IEEE 802.11i/RSN): // bit0 = WPA // bit1 = IEEE 802.11i/RSN (WPA2) (dot11RSNAEnabled) // Note that WPA3 is also configured with bit1 since it uses RSN just like WPA2. // In other words, for WPA3, wpa=2 is used the configuration (and // wpa_key_mgmt=SAE for WPA3-Personal instead of wpa_key_mgmt=WPA-PSK). //wpa=2 For client compatibility therefore: EAPOLv1 (802.1X-2001) should be used by default where WPA is enabled. EAPOLv2 (802.1X-2004) should be used by default where WPA is disabled. To fix this, we can therefore change in the script: set_default eapol_version 0 To the following: set_default eapol_version $((wpa & 1)) This therefore: 1) Sets eapol_version to 1 where WPA has been enabled via wpa bit0 being set. 2) Sets eapol_version to 0 where WPA has been disabled via wpa bit0 being unset. For usual configurations that only have WPA2 enabled, EAPOLv2 is then used. Signed-off-by: Nick Lowe <nick.lowe@gmail.com>
* ramips: unify elecom-*-factory for ELECOM WRC-GHBK2-S/GS/GST devicesINAGAKI Hiroshi2020-12-221-16/+5
| | | | | | | | | Most of Build/elecom-wrc-factory and Build/elecom-wrc-gs-factory are nearly equal, Unify those definitions by using "-N" option of mkhash and splitting the appending text at the end of firmware image for WRC-GS/GST devices. Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com>
* scripts: add -N option to mkhash for printing without newlineINAGAKI Hiroshi2020-12-221-7/+13
| | | | | | Added "-N" option, it allow printing hash(es) without newline. Signed-off-by: INAGAKI Hiroshi <musashino.open@gmail.com>
* ath79: fix Sitecom WLR-8100 lan port ledsDavide Fioravanti2020-12-221-3/+3
| | | | | | | | | Incorrect values were used for the switch initialization causing the lan port leds to not light up in case of 10Mb or 100Mb connections. This commit fixes this problem and removes unused values. Signed-off-by: Davide Fioravanti <pantanastyle@gmail.com>
* kernel: move some disabled symbols to genericAleksander Jan Bajkowski2020-12-2227-57/+5
| | | | | | Move some disabled config options found in lantiq target to generic. Signed-off-by: Aleksander Jan Bajkowski <A.Bajkowski@stud.elka.pw.edu.pl>
* kernel: bump 5.4 to 5.4.85John Audia2020-12-2223-47/+47
| | | | | | | | | | | | | All modifications made by update_kernel.sh run in a fresh clone without any existing toolchains. Build system: x86_64 Build-tested: ipq806x/R7800, ath79/generic, bcm27xx/bcm2711 Run-tested: ipq806x/R7800 No dmesg regressions, everything functional Signed-off-by: John Audia <graysky@archlinux.us>
* Revert "mvebu: fix initramfs/kernel image for CZNIC Turris Omnia"Petr Štetiar2020-12-221-1/+2
| | | | | | | | | | | This reverts commit e401a2a42e6d7c892e1cf7d765fa5ec9b2db3fb3 as at least two users had reported breakage on their devices. References: https://lists.infradead.org/pipermail/openwrt-devel/2020-December/032837.html References: https://github.com/openwrt/openwrt/commit/e401a2a42e6d7c892e1cf7d765fa5ec9b2db3fb3#commitcomment-45189788 Cc: Karel Kočí <karel.koci@nic.cz> Reported-by: Magnus Kroken <mkroken@gmail.com> Signed-off-by: Petr Štetiar <ynezz@true.cz>
* mac80211: fix MAC address allocations when local bit set on base addrPaul Fertser2020-12-221-2/+2
| | | | | | | | | | | | | | | Testing with hwsim reveals two problems: 1. phyX/addresses has two addresses and mac80211_get_addr keeps returning the last one when asked for more; 2. The base address has the local bit set and the operation unsets it. Fix both. Fixes: 866790fd827cb0187353cdf484eb46a9b38fb6ba Reported-by: Zero_Chaos Signed-off-by: Paul Fertser <fercerpav@gmail.com>
* mvebu: gpio-mvebu IRQ index error kernel backportDaniel González Cabanelas2020-12-221-0/+33
| | | | | | | | | Backport the upstream kernel fix 525b0858ff to get rid of the kernel messages: mvebu-gpio xxxxxx.gpio: IRQ index 3 not found Signed-off-by: Daniel González Cabanelas <dgcbueu@gmail.com>
* ipq806x: Fix gmac1 (WAN) on Netgear D7800Peter Cardoe2020-12-221-1/+5
| | | | | | | | | | | | | The WAN port on Netgear D7800 is unable to connect to an ISP when the link to the modem is running at less than 1GB. This patch fixes the issue by removing the phy-handle definition and replacing it with a fixed-link definition The WAN port is then able to connect to a modem via a link running at 100Mbs or 10Mbs Fixes: FS#3086 Signed-off-by: Peter Cardoe <peter@cardoe.co.uk>
* mvebu: LS421DE: use the RTC as a wakeup sourceDaniel González Cabanelas2020-12-221-0/+1
| | | | | | | | | | | | | | | | | | | The Buffalo Linkstation LS421DE comes with a Ricoh RS5C372A real time clock. This RTC has the INTRA pin connected to the power management circuit, allowing to wake up the device from the power off state when an alarm is scheduled. Add the "wakeup-source" property in the RTC dts node to allow the use of the alarm. Example of use, the device is powered off and it comes to life after 5 minutes: echo $(expr $(date '+%s') + 60 * 5) > /sys/class/rtc/rtc0/wakealarm poweroff This feature isn't available in the stock firmware. Signed-off-by: Daniel González Cabanelas <dgcbueu@gmail.com>
* kernel: rtc: rs5c372: fix alarm supportDaniel González Cabanelas2020-12-222-0/+168
| | | | | | | | | | | | | | The rs5c372 RTC driver has alarm support, but it can't be enabled and only can handle 24 hours in the future spite the chip is 1 week capable. Provide these two patches: - Support alarms up to 1 week - Let the wakealarm to be used as a wakeup source This patch makes the alarm wakeup feature to be available in the Buffallo Linkstation LS421DE (mvebu target) and should also work with any other device if the hardware has the proper capability. Signed-off-by: Daniel González Cabanelas <dgcbueu@gmail.com>
* base-files: flush kernel memory cache during sysupgradeHannu Nyman2020-12-222-0/+2
| | | | | | | | | | | Flush kernel memory caches during sysupgrade in order to mitigate the impact from memory consumption spikes in low-RAM devices. This may help to prevent sysupgrade causing a reboot before the actual flashing starts. Signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>
* mvebu: LS421DE: fix low performance issueDaniel González Cabanelas2020-12-221-0/+4
| | | | | | | | | | The Buffalo Linkstation LS421DE isn't able to enable the Level 2 cache (AKA Aurora cache). As of result of this, the throughput is about half of the expected, e.g when doing network data transfers. Fix it by adding the broken-idle property in the coherency fabric node. Signed-off-by: Daniel González Cabanelas <dgcbueu@gmail.com>
* tegra: image: remove unnecessary assumptions from bootscriptTomasz Maciej Nowak2020-12-222-4/+3
| | | | | | | | | | | 'bootz' expects gziped kernel image anyway, so hard-code it to zImage, and remove root path from 'load' commands, by default the files are searched in root directory. This will make the bootscript static, so the command which modified it when image was created can now be removed. Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
* tegra: image: make bootscript standaloneTomasz Maciej Nowak2020-12-222-5/+4
| | | | | | | Don't hard-code the PTUUID, use U-Boot commands to determine it, as some partitioning tools could rewrite PTUUID when modifying partitions. Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
* uboot-tegra: bump to 2020.04Tomasz Maciej Nowak2020-12-221-2/+2
| | | | | | | | | This fixes error when host GCC >= 10. /usr/bin/ld: scripts/dtc/dtc-parser.tab.o:(.bss+0x10): multiple definition of `yylloc'; scripts/dtc/dtc-lexer.lex.o:(.bss+0x0): first defined here collect2: error: ld returned 1 exit status Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
* tegra: sysupgrade: remove check for number of argumentsTomasz Maciej Nowak2020-12-221-2/+0
| | | | | | | | This chunk got mistakenly removed from 30c95c4, since the get_image_dd evaluates only first agument, so that check is useless. Fixes: 30c95c4 ("tegra: sysupgrade: use get_image_dd wrapper") Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
* tcpdump: fix pcap-config issuesRosen Penev2020-12-222-115/+4
| | | | | | | | | | | | | | The patch removes a libpcap check to avoid a problem with libpcap. Fix libpcap instead. Modernize Makefile: Use a normal autoconf bool instead of checking for CONFIG_IPV6. Remove old configure and MAKE_FLAGS hacks. Removing them results in compilation continuing to work without a problem. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* libpcap: fix pcap-configRosen Penev2020-12-221-2/+10
| | | | | | | | | | | | | pcap-config as installed is using OS paths instead of OpenWrt ones. Take fix from libpng and adjust as needed. This problem seems to occur on Arch Linux and not on Debian/Fedora based distros. No idea why. Remove CMAKE_INSTALL as there is now an InstallDev section. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* target/linux: disable CONFIG_CIFS_STATSRosen Penev2020-12-221-1/+1
| | | | | | | | CIFS_STATS is a debugging option. It is not really useful for end users Signed-off-by: Rosen Penev <rosenp@gmail.com> [fixed missing config-4.19 file] Signed-off-by: Petr Štetiar <ynezz@true.cz>
* fs/cifs: update module dependenciesRosen Penev2020-12-221-7/+10
| | | | | | | | | | | 0fdfef9aa7ee68ddd508aef7c98630cfc054f8d6 upstream removed CIFS_SMB311. Kernels 4.19 and above do not have it. Currently only kernels 4.19 and 5.4 are in the tree. The Kconfig file in the kernel has more selection that what is in here. Add the rest and reorder based on upstream ordering. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* pcre: fix paths in config fileRosen Penev2020-12-221-1/+2
| | | | | | | | | The paths are pointing to OS paths, not OpenWrt ones. Use SED line from libpng to fix and adjust accordingly. This may allow certain packages that use the config file to pick up pcre. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* nettle: update to 3.6Rosen Penev2020-12-221-6/+6
| | | | | | | | | | | Updated ABI_VERSION. Switched PKG_BUILD_PARALLEL on as there seems to be no issue anymore. I can't find any information about why it was turned off. Fixed license information. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* binutils: update to 2.35.1Rosen Penev2020-12-221-2/+2
| | | | | | Fixes compilation with musl 1.2.x. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* elfutils: update to 0.180Rosen Penev2020-12-223-6/+6
| | | | | | Refreshed patches. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* tools/pkgconf: update to 1.7.3Rosen Penev2020-12-222-38/+2
| | | | | | Remove upstreamed patch. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* ccache: update to 4.1Rosen Penev2020-12-222-18/+15
| | | | | | | | | | Upstream switched to building with CMake. Adjust accordingly. Reapplied patch as upstream changed the file format. Added HOST_BUILD_PARALLEL for faster compilation. Signed-off-by: Rosen Penev <rosenp@gmail.com>
* hostapd: parse skip_inactivity_poll optionNadim Atiya2020-12-222-3/+5
| | | | | | | | | | | | hostapd.sh does not parse skip_inactivity_poll boolean from /etc/config/wireless despite being mentioned in the documentation [1]. This change fixes this, and by default sets its value to 0 [1]. [1] https://openwrt.org/docs/guide-user/network/wifi/basic Signed-off-by: Nadim Atiya <nadim.atiya@gmail.com> Signed-off-by: Daniel Golle <daniel@makrotopia.org> [fix and reformat commit message, make patch apply]
* include/cmake.mk: Fix the bug in ccache supportHirokazu MORIKAWA2020-12-211-18/+14
| | | | | | | | | | Fix the bug in ccache support due to this change. https://github.com/openwrt/openwrt/commit/bfc433efd4a0c6875a92981d1bd2a5e3e60c61c6 The new cmake uses this for the wrapper. https://cmake.org/cmake/help/v3.19/prop_tgt/LANG_COMPILER_LAUNCHER.html Signed-off-by: Hirokazu MORIKAWA <morikw2@gmail.com>
* base-files: upgrade: fix initramfs detectionRobert Marko2020-12-202-2/+2
| | | | | | | | | | | | | | | Commit "initramfs: switch to tmpfs to fix ujail" switched initramfs to now use tmpfs, it causes $(rootfs_type) to now return tmpfs when running initramfs image instead of being empty. This broke initramfs detection which prevents config files from being saved as it does not work from initramfs. So, lets test for $(rootfs_type) returning "tmpfs" instead. Fixes: 7fd3c68 ("initramfs: switch to tmpfs to fix ujail) Signed-off-by: Robert Marko <robimarko@gmail.com>
* treewide: fix initramfs detectionRobert Marko2020-12-202-2/+2
| | | | | | | | | | | | | | | | Commit "initramfs: switch to tmpfs to fix ujail" switched initramfs to now use tmpfs, it causes $(rootfs_type) to now return tmpfs when running initramfs image instead of being empty. This broke initramfs detection which is required so that when installing on MikroTik devices firmware partition would first get erased fully before writing. So, lets test for $(rootfs_type) returning "tmpfs" instead. Fixes: 7fd3c68 ("initramfs: switch to tmpfs to fix ujail) Signed-off-by: Robert Marko <robimarko@gmail.com>
* uhttpd: don't redirect to HTTPS by defaultPetr Štetiar2020-12-201-1/+1
| | | | | | | | | | | | | | | | | | | So we can ship px5g-wolfssl by default in the release image, but still make the HTTPS for LuCI optional. This small change with addition of `CONFIG_PACKAGE_px5g-wolfssl=y` into the buildbot's seed config for the next release should provide optional HTTPS in the next release. Disabling the current default automatic uhttpd's redirect to HTTPS should make the HTTPS optional. That's it, user would either need to switch to HTTPS by manually switching to https:// protocol in the URL or by issuing the following commands to make the HTTPS automatic redirect permanent: $ uci set uhttpd.main.redirect_https=1 $ uci commit uhttpd $ service uhttpd reload Signed-off-by: Petr Štetiar <ynezz@true.cz>
* mvebu: macchiatobin-singleshot: enable heartbeat LEDTomasz Maciej Nowak2020-12-202-1/+66
| | | | | | | | | | | | | | | With board revision 1.3, SolidRun moved the power LED to the middle of the board. In old place of power LED a GPIO controllable heartbeat LED was added. This commit touches only Single Shot variant, because all have revision 1.3. Some boards could be placed in an enclosure, therefore the LED18 is enabled by default, since that'll be the only visible indicator that the board is operating. Ref: http://wiki.macchiatobin.net/tiki-index.php?page=Schematics#Revision1.3_Electrical_Design_Changes Ref: http://macchiatobin.net/wp-content/uploads/2017/11/MACCHIATOBin-rev-1_3.pdf Reported-by: Alexandra Alth <alexandra@alth.de> Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
* mt76: Fix compile against glibcHauke Mehrtens2020-12-192-1/+26
| | | | | | The mt76 test tools did not compile against glibc. Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
* toolchain: Deactivate sanitizer on MIPS and ARCHauke Mehrtens2020-12-191-4/+4
| | | | | | | | | | MIPS 32 bit support for sanitizer was added with GCC 9, MIPS 64 bit and ARC are still not supported in GCC 10. Deactivate them for now and change this when we change the default compiler to GCC 9 or later. Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
* ath79: airtight c-75: use second flash chipTomasz Maciej Nowak2020-12-192-6/+23
| | | | | | | | | | | The flash capacity is divided in two flash chips and currently only first is used. Increase available space for OpenWrt by additional 16 MiB using mtd-concat driver. Because U-Boot might not be able to load kernel image spanned through two flash chips, the size of kernel is limited to space available on first first chip. Cc: Vladimir Georgievsky <vladimir.georgievsky@yahoo.com> Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
* ath79: add support for AirTight C-75Tomasz Maciej Nowak2020-12-193-0/+186
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | AirTight Networks (later renamed to Mojo Networks) C-75 is a dual-band access point, also sold by WatchGuard under name AP320. Specification SoC: Qualcomm Atheros QCA9550 RAM: 128 MiB DDR2 Flash: 2x 16 MiB SPI NOR WIFI: 2.4 GHz 3T3R integrated 5 GHz 3T3R QCA9890 oversized Mini PCIe card Ethernet: 2x 10/100/1000 Mbps QCA8334 port labeled LAN1 is PoE capable (802.3at) USB: 1x 2.0 LEDs: 7x which two are GPIO controlled, four switch controlled, one controlled by wireless driver Buttons: 1x GPIO controlled Serial: RJ-45 port, Cisco pinout baud: 115200, parity: none, flow control: none JTAG: Yes, pins marked J1 on PCB Installation 1. Prepare TFTP server with OpenWrt initramfs-kernel image. 2. Connect to one of LAN ports. 3. Connect to serial port. 4. Power on the device and when prompted to stop autoboot, hit any key. 5. Adjust "ipaddr" and "serverip" addresses in U-Boot environment, use 'setenv' to do that, then run following commands: tftpboot 0x81000000 <openwrt_initramfs-kernel_image_name> bootm 0x81000000 6. Wait about 1 minute for OpenWrt to boot. 7. Transfer OpenWrt sysupgrade image to /tmp directory and flash it with: sysupgrade -n /tmp/<openwrt_sysupgrade_image_name> 8. After flashing, the access point will reboot to OpenWrt. Wait few minutes, until the Power LED stops blinking, then it's ready for configuration. Known issues Green power LED does not work. Additional information The U-Boot fails to initialise ethernet ports correctly when a UART adapter is attached to UART pins (marked J3 on PCB). Cc: Vladimir Georgievsky <vladimir.georgievsky@yahoo.com> Signed-off-by: Tomasz Maciej Nowak <tmn505@gmail.com>
* bcm53xx: use correct case for TP-Link in DEVICE_VENDORMoritz Warning2020-12-191-2/+2
| | | | | | | | Adjust spelling of vendor names to what is used in other places. Signed-off-by: Moritz Warning <moritzwarning@web.de> [improve commit title/message] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* ipq40xx: use upper case for NETGEAR in DEVICE_VENDORMoritz Warning2020-12-191-2/+1
| | | | | | | | | | Adjust spelling of vendor name to what is used in other places. Also move definition in shared section. Signed-off-by: Moritz Warning <moritzwarning@web.de> [improve commit title/message] Signed-off-by: Adrian Schmutzler <freifunk@adrianschmutzler.de>
* lantiq: xrx200: remove redundant stp parametersAleksander Jan Bajkowski2020-12-193-10/+0
| | | | | | | These parameters are the same as in vr9.dtsi. This patch removes redundant parameters. Signed-off-by: Aleksander Jan Bajkowski <A.Bajkowski@stud.elka.pw.edu.pl>
* mac80211: do not drop tx nulldata packets on encrypted linksFelix Fietkau2020-12-181-0/+25
| | | | | | Fixes sending out nulldata probing frames Signed-off-by: Felix Fietkau <nbd@nbd.name>
* tools/libressl: update to 3.3.1Rosen Penev2020-12-181-2/+2
| | | | Signed-off-by: Rosen Penev <rosenp@gmail.com>
* mt76: update to the latest versionFelix Fietkau2020-12-181-3/+3
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | 7f53d68b1af9 mt76: mt7615: add debugfs knob for setting extended local mac addresses 1a2547b27dfc mt76: do not set NEEDS_UNIQUE_STA_ADDR for 7615 and 7915 2741fd071bb7 mt76: mt7915: support 32 station interfaces 709f2cd77810 mt76: mt7915: fix processing txfree events 434940e08233 mt76: mt7915: measure channel noise and report it via survey 236326896589 mt76: mt7615: retry if mt7615_mcu_init returns -EAGAIN b5c593b63f4c mt76: mt7663s: move tx/rx processing in the same txrx workqueue 75157b59ae4e mt76: mt7663s: convert txrx_work to mt76_worker 6dc67b058e2a mt76: mt7663s: disable interrupt during txrx_worker processing b381729626bb mt76: sdio: convert {status/net}_work to mt76_worker 9cb12f7042cc mt76: mt7915: fix DRR sta bss group index 75977a85e844 mt76: mt7915: disable OFDMA/MU-MIMO UL 6cdebe805862 mt76: rename __mt76_mcu_send_msg to mt76_mcu_send_msg eb9afae96b65 mt76: rename __mt76_mcu_skb_send_msg to mt76_mcu_skb_send_msg 8c73f3b15ada mt76: implement .mcu_parse_response in struct mt76_mcu_ops fcfbb046c2f3 mt76: move mcu timeout handling to .mcu_parse_response 477caa196ffe mt76: move waiting and locking out of mcu_ops->mcu_skb_send_msg a4d71501bad6 mt76: make mcu_ops->mcu_send_msg optional 80c310c59ad1 mt76: mt7603: switch to .mcu_skb_send_msg 082b26181013 mt76: implement functions to get the response skb for MCU calls ee40800df2e4 mt76: mt7915: move eeprom parsing out of mt7915_mcu_parse_response d33943baac47 mt76: mt7915: query station rx rate from firmware b8874e8756d9 mt76: add back the SUPPORTS_REORDERING_BUFFER flag 633ae5961db6 mt76: mt7615: enable beacon filtering by default for offload fw 9a203fea3540 mt76: mt7615: introduce quota debugfs node for mt7663s f9ae638af7e2 mt76: mt7663s: get rid of mt7663s_sta_add 1a5758d894d0 mt76: mt7663s: fix a possible ple quota underflow dea10c03316f mt76: sdio: get rid of sched.lock eb4c09957938 mt76: mt7915: set fops_sta_stats.owner to THIS_MODULE 594890b11155 mt76: mt7915: update ppe threshold 8884a5def518 mt76: mt7915: rename mt7915_mcu_get_rate_info to mt7915_mcu_get_tx_rate 33b89f4a1bf4 mt76: set fops_tx_stats.owner to THIS_MODULE 4d019c9672ec sync with upstream changes 35e3cd1db479 mt76: mt7603: fix ED/CCA monitoring with single-stream devices 4f9f79b085b1 wireless: mt76: convert tasklets to use new tasklet_setup() API 20e8cf935ed0 mt76: dma: fix possible deadlock running mt76_dma_cleanup 36089a655f58 mt76: mt7915: fix sparse warning cast from restricted __le16 68c4eedafd61 mt76: fix memory leak if device probing fails 9a1a0a4dec71 mt76: mt7603: add additional EEPROM chip ID 01b943295719 mt76: move mt76_mcu_send_firmware in common module 0aee4999902a mt76: mt7663s: introduce WoW support via GPIO 79ebad117325 mt76: switch to wep sw crypto for mt7615/mt7915 af139725193a mt76: fix tkip configuration for mt7615/7663 devices 664e66b35c0b mt76: mt7615: run key configuration in mt7615_set_key for usb/sdio devices f675358267d6 mt76: mt76u: rely on woker APIs for rx work b9f9c16cb1bd mt76: mt76u: use dedicated thread for status work cdeb1b29cd15 mt76: testmode: switch ib and wb rssi to array type for per-antenna report 0a898c0549b6 mt76: testmode: add snr attribute in rx statistics 3ea9a0433bcc mt76: testmode: add tx_rate_stbc parameter 73427ebbbd27 mt76: testmode: add support for LTF and GI combinations for HE mode 88ebccfe8a39 mt76: mt7915: fix tx rate related fields in tx descriptor 9909c0551e4c mt76: testmode: add support for HE rate modes 03ed0909f922 mt76: mt7915: implement testmode tx support 0aa696834a9c mt76: mt7915: implement testmode rx support 5ed3a34b46ce mt76: mt7915: add support to set txpower in testmode f86361654e94 mt76: mt7915: add support to set tx frequency offset in testmode 64a765be750a mt76: mt7915: make mt7915_eeprom_read static 9b48c13b52f7 mt76: mt7915: use BIT_ULL for omac_idx 27227fd57ea7 mt76: mt7915: remove unused mt7915_mcu_bss_sync_tlv() cd795267612d mt76: mt7615: support 16 interfaces 82da525ad0c8 mt76: mt7615: refactor usb/sdio rate code b9a50da503ad mt76: mt7915: rely on eeprom definitions c79d18723df0 mt76: move mt76_init_tx_queue in common code b0b221e91445 mt76: sdio: introduce mt76s_alloc_tx_queue caba5a99e5ae mt76: sdio: rely on mt76_queue in mt76s_process_tx_queue signature 3ed4aad81ce9 mt76: mt7663s: rely on mt76_queue in mt7663s_tx_run_queue signature 216cf8b28579 mt76: dma: rely on mt76_queue in mt76_dma_tx_cleanup signature 0f9350bef1b5 mt76: rely on mt76_queue in tx_queue_skb signature 8932975be066 mt76: introduce mt76_init_mcu_queue utility routine b0eb7edcc624 mt76: rely on mt76_queue in tx_queue_skb_raw signature ccd62467d0f3 mt76: move mcu queues to mt76_dev q_mcu array 2e217fb9e962 mt76: move tx hw data queues in mt76_phy 576647f2ec6a mt76: mt7915: fix endian issues e881fd67c718 mt76: move band capabilities in mt76_phy c728cecd7b77 mt76: rely on mt76_phy in mt76_init_sband_2g and mt76_init_sband_5g 231ef27697f9 mt76: move band allocation in mt76_register_phy 8aa24c91b13b mt76: move hw mac_addr in mt76_phy b436da4d9d92 mt76: mt7915: introduce dbdc support 1e34a02c2dcb mt76: mt7915: get rid of dbdc debugfs knob d8e681bd3268 mt76: mt7615: fix rdd mcu cmd endianness 19c9e277eff6 mt76: mt7915: fix memory leak in mt7915_mcu_get_rx_rate() e361b6a71e4b mt76: improve tx queue stop/wake fb24e5f2305b mt76: mt7915: stop queues when running out of tx tokens 066cc441eb8f mt76: attempt to free up more room when filling the tx queue 93c806a34ec2 mt7915: fix minor issues in the token queue blocking change c017e329a326 mt76: mt7915: ensure that init work completes before starting the device 9e9da427b8a6 mt76: mt7915: fix polling firmware-own status 5cd805ddfb25 mt76: add more conditions for stopping tx scheduling aa893c73bf85 mt76: mt7915; increase txq memory limit for non-DBDC 7915 cards to 32 MiB e44b7c91070e mt76: skip queue stop/wake, rely entirely on txq scheduling 6c6a5c59c101 mt76: mt7915: do not set DRR group for stations 510cb5be1bf7 mt76: usb: remove wake logic in mt76u_status_worker 34f318a25421 mt76: sdio: remove wake logic in mt76s_process_tx_queue 4a90fdf6105e mt76: mt76s: fix NULL pointer dereference in mt76s_process_tx_queue Signed-off-by: Felix Fietkau <nbd@nbd.name>
* kernel: improve skb hash on the mtk ethernet driverFelix Fietkau2020-12-185-25/+35
| | | | | | | | The PPE only provides a 14 bit hash, however many uses of the skb hash expect the hash to use the full 32 bit range. Use jhash to extend the hash to the full size Signed-off-by: Felix Fietkau <nbd@nbd.name>
generated by cgit v1.2.3 (git 2.25.1) at 2024-06-29 08:47:33 +0000