Despite Westwood's theoretical advantages, in nearly
every benchmark we ran last year, TCP cubic won, whether it be
on correct RTT estimates, amount of buffering, responsiveness,
etc. on current hardware and software designs.
(both need timestamps on to work well, besides)
TCP cubic is better maintained and understood than westwood,
also.
While a scenario where westwood would win possibly exists,
there is too much buffering in the wifi stack in particular
at present, to see any improvement.
If you wish to exercise various TCPs under contention,
the current svn head of netperf (2.6) has options to switch
congestion control agorithms on the fly, as does iperf.
SVN-Revision: 32514
Detect and apply MAC address of a Fonera 2.0n (ramips)
Signed-off-by: Malte Forkel <malte.forkel@berlin.de>
Signed-off-by: Vasilis Tsiligiannis <b_tsiligiannis@silverton.gr>
SVN-Revision: 32490
Hardware filtering must always be enabled as long as there is an Ethernet
device registered, and use device tree for setting the link activity and
buffer shifting enable/disable bit.
SVN-Revision: 32486
This target currently only supports Moschip's MCS8140 SoC, but support
for other chips in the same family (MCS8142, MCS8144) will be easy to add.
Target support is entirely using Device Tree for probing peripherals.
Drivers support include:
- PCI
- USB 1 & 2
- watchdog
- random number generator
- UART
- timer
- internal Ethernet PHY
- Ethernet MAC core
Support for the following boards is included using Device Tree
- Devolo dLAN USB Extender
- Tigal RBT-832
SVN-Revision: 32462
This patch helps properly detect a WNDRMAC device. Before this
patch the model is detected as "NETGEAR ?????????N".
Signed-off-by: Roman A. aka BasicXP <x12ozmouse@ya.ru>
SVN-Revision: 32453
This adds kmod-leds-wndr3700-usb package to a default build for
WNDR3700, WNDR3800 and WNDRMAC in order to support the green USB LED.
Signed-off-by: Roman A. aka BasicXP <x12ozmouse@ya.ru>
SVN-Revision: 32452
This patch adds support for the Korean made Petatel PSR-680W Wireless CDMA Router.
The platform is based on Ralink RT3052.
http://wiki.openwrt.org/toh/petatel/psr-680w
Signed-off-by: Dmitry Shmygov <shmygov at rambler.ru>
SVN-Revision: 32450
This patch enables few extra kernel options for the kvm_guest subtarget:
- Tickless kernel to avoid timer ticks in idle guests, reduces CPU usage
- Enable paravirtualization steal time support
Signed-off-by: Stijn Tintel <stijn@linux-ipv6.be>
SVN-Revision: 32436
BCM6338 and BCM6338 have their MSG_CONTROL register width of 8-bits instead of
16-bits. We were previously using a 16-bits write which corrupted the first
byte of the TX FIFO. Also the message type was always set to Full-duplex even
in the case of half-duplex messages.
SVN-Revision: 32409
Merge most fixes from #8677:
- add basic hardware detection of Livebox Blue 5g revisions
- register leds and gpio buttons
- fix boot address location
- properly parse mac addresses
Plus some more fixes:
- make board_livebox.c more in line with board_bcm963xx.c
SVN-Revision: 32087
Some new board definitions were added within the #ifdef CONFIG_CPU_6338
section while some other were added within #ifdef CONFIG_CPU_6345, fix this
SVN-Revision: 32060
Build images for 96328avng and 963281TAN (untested) for three different
flash sizes: 4MB, 8MB, and 16MB.
Known issues:
* Button setup unknown, thus no fail-safe available.
* All other BCM6328 and BCM63XX issues
SVN-Revision: 32004
Add led devices for D-Link DIR-300-B1 WAN LED.
Note that the GPIO state is also ANDed with the esw switch port 4
LED state, which is why I've set the amber LED to default-on.
Closes: https://dev.openwrt.org/ticket/11326
Signed-off-by: Tobias Diedrich <ranma+openwrt@tdiedrich.de>
[ juhosg:
- fix comment style in mach-dir-300-revb.c,
- remove the PPP specific LED setup, not everyone uses a PPP connection
on the WAN interface]
SVN-Revision: 31989
WIFI eeprom:
As discussed, it is impossible for nand and spi flash platforms to have the eeprom data available from mtd. I suggested to load the eeprom from user-space. I've looked into regular firmware loading but this is only possible when using modules. I've created a sysfs entry that allows reading and writing the eeprom data to the platform data. After loading the eeprom data I rely on pci-hotplug support to disable the bogus pci device and rescan the bus (with fixups and all). Because hotplug is not available, an init script is created that performs the copy from mtd to platform data. I think it is best to eventually move the sysfs functions to dev_wifi_athxk.c file, this would get rid of the external to the ath9k platform data.
SPI flash:
It seems that the spi-xway driver is not really working. It causes my kernel to crash in all sorts of ways. I added to bitbang SPI to be able to the calibration data mention above.
I've kept the original mtd partitioning that Netgear uses.
Buttons/LED:
Both buttons are working properly. Two leds are not enabled: The red internet led is connected to the pci_gnt1 pin... I can't disable it because then DMA stops working.
The green wifi led is connected to an unknown atheros gpio.
Signed-off-by: Pieter Voorthuijsen <p.voorthuijsen at gmail.com>
SVN-Revision: 31910
moves the linux partition, removes the config partition and the
buttons settings have small adjustments.
Signed-off-by: Luka Perkov <openwrt@lukaperkov.net>
SVN-Revision: 31909
Build images for D-Link DSL-274XB revision F1. Flashing
through original web UI as well as CFE works.
Known issues:
* wifi led inverted (on when wifi is off and vice cersa)
* same issues as generic bcm6328 support
SVN-Revision: 31885
BCM6328 support is in a usable state, with the following issues
* internal switch port not configurable
* external phys not supported
* broadcom wifis will probably not work correctly
* and all other generic bcm63xx issues (so yes, still no ADSL)
SVN-Revision: 31883
bcm_enetsw_probe did not properly reset port states, resulting in
connected ports staying disabled after bringing up the interface the
second time.
SVN-Revision: 31881
Support for TP-Link TL-MR3040. Power LED not working - always on red, after
start should be green.
Signed-off-by: Cezary Jackiewicz <cezary.jackiewicz@gmail.com>
SVN-Revision: 31845
Add byte queue limits support to net/ethernet/ramips_main.c
"Byte queue limits are a mechanism to limit the size of the transmit
hardware queue on a NIC by number of bytes. The goal of these byte
limits is too reduce latency (HOL blocking) caused by excessive
queuing in hardware (aka buffer bloat) without sacrificing
throughput."
Signed-off-by: Tobias Diedrich <ranma+openwrt@tdiedrich.de>
SVN-Revision: 31844
Sysupgrade works just fine on my Sitecom WL-351 after adding this
oneliner.
Signed-off-by: Tobias Diedrich <ranma+openwrt@tdiedrich.de>
SVN-Revision: 31843
