This adds a patch that introduces airtime fairness scheduling to ath9k,
which can significantly improve network efficiency in mixed-rate
environments.
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
The patch commit states:
"It's possible to make scanning consume almost arbitrary amounts
of memory, e.g. by sending beacon frames with random BSSIDs at
high rates while somebody is scanning.
Limit the number of BSS table entries we're willing to cache to
1000, limiting maximum memory usage to maybe 4-5MB, but lower
in practice - that would be the case for having both full-sized
beacon and probe response frames for each entry; this seems not
possible in practice, so a limit of 1000 entries will likely be
closer to 0.5 MB."
Signed-off-by: Johannes Berg <johannes.berg@intel.com>"
This patch was added in 4.4.36. But because LEDE backports
cfg80211, mac80211 and the wifi drivers separately, it needs
to be added manually for now. It can be dropped later as it
will be part of the next mac80211 refresh.
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
This works around deadlock and/or memory corruption during
firmware crash and improves ability to configure number of
tids in firmware.
Signed-off-by: Ben Greear <greearb@candelatech.com>
Currently, installing kmod-sdhci fails with "sdhci is already loaded" since
"sdhci.ko" is inserted explicitely first, the implicitely loaded again when
"sdhci-pltfm.ko" is inserted as the latter depends on the former.
Remove the explicit autoprobe of "sdhci.ko" to fix the postinstall script.
Acked-by: John Crispin <john@phrozen.org>
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
This will avoid loading it in the default configuration, which reduces
image size a bit, and (more importantly) improves performance by
avoiding some unnecessary netfilter hooks
Signed-off-by: Felix Fietkau <nbd@nbd.name>
this kernel module currently does not set submenu.
Fix this by adding it to the "Others" submenu
Signed-off-by: Alberto Bursi <alberto.bursi@outlook.it>
Track upstream changes, incl changes in packet overhead accounting
(automatically taking care of linux' packet sizing knowledge),
improvements to triple isolated DRR handling (new flow dominance),
statistics tweak & allow more packet drops in stressed conditions.
Under tests this has significantly improved latency control under
'many flows to one' scenarious as is typical of bittorrent and MS
Windows update.
I also restored 'DSCP washing' functionality in my repo which follows
upstream closely (like a hawk!) with tc keywords 'wash/nowash'. This
allows cake to limit/control packets in bands determined by a packet's
DSCP but to clear those DSCP bits on qdisc egress. This functionality
was originally removed as part of an attempt to push cake into the
kernel, which hasn't actually happened as yet.
A matching commit is required to iproute2/tc to support the new overhead
handling, keyword changes as well as the 'wash/nowash' tweak.
Signed-off-by: Kevin Darbyshire-Bryant <kevin@darbyshire-bryant.me.uk>
This just adds the kmods for these kernel modules.
This is found on some Lantiq / Intel reference boards.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
These properties allow overriding the settings from the EEPROM
which indicate whether a band is enabled or not.
Setting this property is only needed when the RF circuit does not
support the 2.4GHz or 5GHz band while it is enabled nevertheless in the
EEPROM.
These patches will be replaced with a future upstream version which
will introduces an ieee80211 device tree property to disable bands.
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
There are two types of swapping the EEPROM data in the ath9k driver.
Before this series one type of swapping could not be used without the
other.
The first type of swapping looks at the "magic bytes" at the start of
the EEPROM data and performs swab16 on the EEPROM contents if needed.
The second type of swapping is EEPROM format specific and swaps
specific fields within the EEPROM itself (swab16, swab32 - depends on
the EEPROM format).
With this series the second part now looks at the EEPMISC register
inside the EEPROM, which uses a bit to indicate if the EEPROM data
is Big Endian (this is also done by the FreeBSD kernel).
This has a nice advantage: currently there are some out-of-tree hacks
(in OpenWrt and LEDE) where the EEPROM has a Big Endian header on a
Big Endian system (= no swab16 is performed) but the EEPROM itself
indicates that it's data is Little Endian. Until now the out-of-tree
code simply did a swab16 before passing the data to ath9k, so ath9k
first did the swab16 - this also enabled the format specific swapping.
These out-of-tree hacks are still working with the new logic, but it
is recommended to remove them. This implementation is based on a
discussion with Arnd Bergmann who raised concerns about the
robustness and portability of the swapping logic in the original OF
support patch review, see [0].
After a second round of patches (= v1 of this series) neither Arnd
Bergmann nor I were really happy with the complexity of the EEPROM
swapping logic. Based on a discussion (see [1] and [2]) we decided
that ath9k should use a defined format (specifying the endianness
of the data - I went with __le16 and __le32) when accessing the
EEPROM fields. A benefit of this is that we enable the EEPMISC based
swapping logic by default, just like the FreeBSD driver, see [3]. On
the devices which I have tested (see below) ath9k now works without
having to specify the "endian_check" field in ath9k_platform_data (or
a similar logic which could provide this via devicetree) as ath9k now
detects the endianness automatically. Only EEPROMs which are mangled
by some out-of-tree code still need the endian_check flag (or one can
simply remove that mangling from the out-of-tree code).
[0] http://www.spinics.net/lists/linux-wireless/msg152634.html
[1] https://marc.info/?l=linux-wireless&m=147250597503174&w=2
[2] https://marc.info/?l=linux-wireless&m=147254388611344&w=2
[3] 50719b56d9/sys/dev/ath/ath_hal/ah_eeprom_9287.c (L351)
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
These patches add support for configuring ath9k based devices via
devicetree. This was tested on PCI(e) based devices. This should work
for AHB based devices as well (adding more AHB specific properties may
still be needed) as soon as the ath79 platform is ready to populate the
ath9k wmac via devicetree.
This patchset was accepted upstream, more information can be found on
the linux-wireless list:
https://www.spinics.net/lists/linux-wireless/msg155474.html
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
The ESP algorithms in CBC mode require echainiv, so have kmod-ipsec
depend on kmod-crypto-echainiv.
See upstream commit 32b6170ca59ccf07d0e394561e54b2cd9726038c.
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
TI wl18xx and wl12xx are Wi-Fi/Bluetooth combo modules
that could be found on different existing boards.
But it is possible to get those modules as a separate
component and use with existing boards as well as
new boards equipped with either module may appear so we
remove dependency on OMAP instead we add dependency on MMC
because this Wi-Fi module uses SDIO interface.
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
Cc: Jo-Philipp Wich <jo@mein.io>
Cc: Felix Fietkau <nbd@nbd.name>
Cc: Imre Kaloz <kaloz@openwrt.org>
Reviewed-by: Florian Fainelli <f.fainelli@gmail.com>
Without setting the HSR to the selected channel, the WLAN of the UAP
Outdoor+ will exhibit high packet loss in RX.
Based-on-patch-by: Stefan Rompf <stefan@loplof.de>
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Backport upstream accepted patch which allows to override the EEPROM
mac address with one from device tree.
Signed-off-by: Mathias Kresin <dev@kresin.me>
The patch 615-rt2x00-fix_20mhz_clk.patch fixes code introduced by
611-rt2x00-rf_vals-rt3352-xtal20.patch and makes the the platform data
property clk_is_20mhz obsolete.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Most of the lantiq devices with ralink wifi have the EEPROM stored
in big endian byte order in flash, but the driver expects the EEPROM to
be in little endian.
Signed-off-by: Mathias Kresin <dev@kresin.me>
The ralink,mtd-eeprom device tree property doesn't require the
ralink,eeprom property to work.
Rework the error handling and user notification as well. Do not log an
error if the mtd-eeprom parameter isn't used. It could be intentional
and should not scare the user.
Check if the number of bytes read from the mtd devices matches the
requested number of bytes.
In case of an mtd read error, give a hint to the user which partition
was tried to read from.
In case everything is fine, notify the user as well.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Call the rt2x00lib_get_eeprom_file_name only once and from the function
where the EEPROM filename is required.
Error only out if an EEPROM file is mandatory. Use the
REQUIRE_EEPROM_FILE bit to determine if it is mandatory.
Do not set the REQUIRE_EEPROM_FILE bit while requesting an EEPROM file.
It should be (and is) set before requesting an EEPROM file.
Do not redirect users to upstream while using a function of a custom
patch.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Package the driver for the DS1374 chip present on the T4240RDB.
Signed-off-by: Florian Larysch <fl@n621.de>
Signed-off-by: Felix Fietkau <nbd@nbd.name> [fix missing symbol]
FS#256
Fix audio not working due to unset SND_PCM_TIMER. CONFIG_SND_PCM_TIMER
is required for ALSA to work (at least for some audio devices),
otherwise applications using sound may fail. Can be reproduced by
installing "alsa-utils-tests" and running "speaker-test":
...
ALSA lib pcm_direct.c:1159:(snd1_pcm_direct_initialize_poll_fd) unable to open timer 'hw:CLASS=3,SCLASS=0,CARD=0,DEV=0,SUBDEV=0'
ALSA lib pcm_dmix.c:1106:(snd_pcm_dmix_open) unable to initialize poll_fd
Playback open error: -19,No such device
The kernel size does not increase, only the size of the kernel module
increases.
Signed-off-by: Dmitry Sutyagin <f3flight@gmail.com>
This reverts commit efd9dec319.
ath10k can take a long time to probe, long enough for netifd to fail to
initialize already configured wireless devices
Signed-off-by: Felix Fietkau <nbd@nbd.name>
LEDE no longer requires all PHYs to be initialized to
create the configuration files during bootup. This patch
removes the now obsolete ath10k patch.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Currently, the wifi detection script is executed as part of
the (early) boot process. Pluggable wifi USB devices, which
are inserted at a later time are not automatically
detected and therefore they don't show up in LuCI.
A user has to deal with wifi detection manually, or restart
the router.
However, the current "sleep 1" window - which the boot
process waits for wifi devices to "settle down" - is too
short to detect wifi devices for some routers anyway.
For example, this can happen with USB WLAN devices on the
WNDR4700. This is because the usb controller needs to load
its firmware from UBI and initialize, before it can operate.
The issue can be seen on a BT HomeHub 5A as well as soon as
the caldata are on an ubi volume. This is because the ath9k
card has to be initialized by owl-loader first. Which has to
wait for the firmware extraction script to retrieve the pci
initialization values inside the caldata.
This patch moves the wifi configuration to hotplug scripts.
For mac80211, the wifi configuration will now automatically
run any time a "ieee80211" device is added. Likewise
broadcom-wl's script checks for new "net" devices which
have the "wl$NUMBER" moniker.
Issues with spawning multiple interface configuration - in
case the detection script is run concurrently - have been
resolved by using a named section for the initial
configuration. Concurrent configuration scripts will now
simply overwrite the same existing configuration.
A workaround which preserves the "sleep 1" window for just
the first boot has been added. This allows the existing
brcm47xx boot and mvebu uci-default scripts to correctly
setup the initial mac addresses and regulatory domain.
And finally, the patch renames the "wifi detect" into
"wifi config". As the script no longer produces any output
that has to be redirected or appended to the configuration
file.
Thanks to Martin Blumenstingl for helping with the implementation
and testing of the patch.
Acked-by: Jo-Philipp Wich <jo@mein.io>
Signed-off-by: Mathias Kresin <dev@kresin.me>
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
Previously, wifi detect simply dumped its generated wireless
configuration to STDOUT. A second step was needed to append
the configuration to /etc/config/wireless (or create it, if
it didn't exist).
With this patch, The wifi detection script will now use uci
to update the wireless configuration directly.
This patch also makes the initially created wifi-iface a
named section ('default_radio$X' for mac80211 and
'default_wl$X' for broadcom). With this change, uci will
not print the cfgHASH to STDOUT (which would now corrupt
the wireless configuration). It will also prevent adding
duplicated wifi interface configurations, if the wifi
configuration is run concurrently.
Signed-off-by: Christian Lamparter <chunkeey@gmail.com>
(required not-distributable firmware blob - dump it by yourself from original firmware)
Signed-off-by: Eddi De Pieri <eddi@depieri.net>
(cherry picked from commit 47b1ff965b0cb57013b40fbe2bcd7f3c6eb6b606)
Modified after cherry-pick:
FW_MD5SUM in Makefile
Signed-off-by: Stefan Koch <stefan.koch10@gmail.com>
(required not-distributable firmware blob - dump it by yourself from original firmware)
Signed-off-by: Eddi De Pieri <eddi@depieri.net>
(cherry picked from commit ea9e61b8eb61a2e362a50541f03466dc7d087947)
Signed-off-by: Stefan Koch <stefan.koch10@gmail.com>
Adds Sebastian's 160Mhz support (un-tested), remove DMA32 change that
broke some x86 systems, allow setting 10.1 CT firmware keepalive watchdog
timeout, support QCA 9887 hardware, and some other tweaks.
Signed-off-by: Ben Greear <greearb@candelatech.com>
This is upstream alternative for LEDE's ledtrig-usbdev. It's main
advantages are:
1) Support for assigning more than 1 USB port to the LED
2) Setting proper state when activating with device already connected
3) FWIW it's an upstream driver
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
The file to patch doesn't exists. Due to the fact that a src directory
exists, the patch was never applied.
With 28502a9 the behaviour was changed and revealed the issue.
Signed-off-by: Mathias Kresin <dev@kresin.me>
Tested with VDSL on TP-Link WD8970, I see full 1500-byte PPP data
frames, which end up being 1526 byte Ethernet frames (including
Ethernet+VLAN headers) on the wire.
Fixes: FS#210
Signed-off-by: David Woodhouse <dwmw2@infradead.org>
This fixes the following error when mounting a ext4 filesystem
----
[ 166.240000] EXT4-fs (sda1): Cannot load crc32c driver.
----
Signed-off-by: Daniel Danzberger <daniel@dd-wrt.com>
nf_tproxy_core was removed during 3.12 development with kernel commit
fd158d79d33d3c8b693e3e2d8c0e3068d529c2dc. The code was moved
to xt_TPROXY.c.
Fixes FS#212
Signed-off-by: Mathias Kresin <dev@kresin.me>
Adds the latest patches from Jes Sorensen for rtl8xxxu, which improve
rtl8732bu, rtl8192eu and rtl8188eu support.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
* Change git packages to xz
* Update mirror checksums in packages where they are used
* Change a few source tarballs to xz if available upstream
* Remove unused lines in packages we're touching, requested by jow- and blogic
* We're relying more on xz-utils so add official mirror as primary source, master site as secondary.
* Add SHA256 checksums to multiple git tarball packages
Signed-off-by: Daniel Engberg <daniel.engberg.lists@pyret.net>
The kmod-sound-hda-core module attempts to package snd-hda-core.ko which
does not exist in Linux 3.18, therfore only use it for kernels >= 4.1
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
Prior to kernel 4.4, the hda-intel module depends on the hda-controller
utility submodule so bundle it for the older kernel versions.
Signed-off-by: Jo-Philipp Wich <jo@mein.io>
vfat filesystem fails to mount due to missing codepages with
factory-formatted flash drives. Depend on cp437 iso8559-1 and
utf8 nls modules as this covers most factory-formatted vfat
filesystems.
Signed-off-by: Daniel Dickinson <lede@cshore.thecshore.com>
This patch set adds support for PCI Intel HD Audio
sound devices. This is useful for multimedia packages
in the packages feed that one may use to create audio
servers.
Signed-off-by: Daniel Dickinson <lede@cshore.thecshore.com>
The nf_reject_* and nf_nat_masquerade_* modules are moved into the
corresponding kmod-nf- packages. Appropriate dependencies are added to the
kmod-nft- packages.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Since the kernel makefile is using .ONESHELL, we need to add -e to
.SHELLFLAGS so errors are not ignored.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
There is a separate package kmod-crypto-echainiv for echainiv.ko. Selecting
both packages led to a conflict, so remove the file from kmod-crypto-iv.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
This fixes bug that could cause WARNING on every add_key/del_key call.
It also replaces WARNING with a simple message. They may still occur
e.g. on station going out of range and A-MPDU stall in the firmware.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
All these patches are in wireless-drirvers-next. There is support for
hidden SSID, few new devices and many fixes.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
br_netfilter.ko and the corresponding Kconfig symbol are already provided
by kmod-br-netfilter, which is a dependency of kmod-ebtables.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Some devices (like the Cisco Meraki Z1 Cloud Managed Teleworker Gateway)
need to be able to initialize the PCIe wifi device. Normally, this is done
during the early stages of booting linux, because the necessary init code
is read from the memory mapped SPI and passed to pci_enable_ath9k_fixup.
However,this isn't possible for devices which have the init code for the
Atheros chip stored on NAND in an UBI volume. Hence, this module can be
used to initialze the chip when the user-space is ready to extract the
init code.
Martin Blumenstingl made a few fixes and added support for lantiq:
kernel: owl-loader: add support for OWL emulation PCI devices
kernel: owl-loader: don't re-scan the bus when ath9k_pci_fixup failed
kernel: owl-loader: use dev_* instead of pr_* logging functions
kernel: owl-loader: auto-generate the eeprom filename as fallback
kernel: owl-loader: add a debug message when swapping the eeprom data
kernel: owl-loader: add missing newlines in log messages
kernel: owl-loader: add support for the lantiq platform
These patches have been integrated. Thanks!
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Now that we have firmwares separated and brcm2708 being the only target that
actually selects SDIO support, avoid selecting all firmwares by default.
sunxi should select the proper firmwares once SDIO support is enabled and
tested.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Using few packages will allow saving some space by decreasing rootfs
size.
Moreover there are more firmware files that may require packaging and
even more to come later.
This can especially useful now, with per device rootfs.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
clean up usb gadget support:
- rename gadget modules so that they appear together and are easier to
identify as gadget modules
- make usb-lib-composite and usb-gadget hidden as there is no point in
selecting those without gadget drivers that require them as deps
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
Now we have firmwares separated and bcm53xx selecting required ones make
use of it to actually save that rootfs space.
Other targets using brcmfmac (brcm2708 and sunxi) use SDIO interface and
firmware so they don't won't be affected.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Using few packages will allow saving some space by decreasing rootfs
size. Dropping 43602a1 firmware saves 316 580 B. Dropping 4366b1 saves
468 188 B.
Moreover there are more firmware files that may require packaging and
even more to come later (e.g. 4366c0).
This can especially useful now, with per device rootfs.
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Add a dependency on DRIVER_11W_SUPPORT in order to enable the IEEE
802.11w functionality in hostapd.
Signed-off-by: Petko Bordjukov <bordjukov@gmail.com>
Add a patch to backport 5185c91385d73cdf79836eb8548e4726e43ae831
from Linux 4.8 adding USB2380 support to the NET2280 driver and
create an OpenWrt menu option to select it as a module.
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
power save response frames can go through the old tx path, and the tid
needs to be set for sequence numbers to be assigned correctly.
Signed-off-by: Felix Fietkau <nbd@nbd.name>
The ATH9K_STATION_STATISTICS kernel config variable enables some extra
statistics that are useful for debugging (in particular with the airtime
fairness patches enabled). This adds that kernel config when selecting
ath9k debugging.
Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
This uses GFP_DMA32 for firmware swap. Fixes issue on x86-64 with
QCA 9984 chipset when host system does not have vt-d enabled.
Also tested on linksys ea8500 with 9980 chipset.
Signed-off-by: Ben Greear <greearb@candelatech.com>
Without any in-tree users enabled the Kernel's build process doesn't
actually build those modules. Enable some potential in-tree users
during Kernel build, so out-of-tree modules can depend on them.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
With the default priority of 0, the DEU algos would be overlapped
by the generic algos (if available).
To fix this, set the cra_priority of the hardware algos to the
recommended value of 300/400.
Signed-off-by: Martin Schiller <mschiller@tdt.de>
