Age | Commit message (Collapse) | Author |
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32680 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32672 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32670 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32604 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32590 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32580 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
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.
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32514 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32498 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32456 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
that applies to all mips targets
The dcache bug that it works around is a generic issue, not a brcm47xx cache quirk
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32395 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32094 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32085 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32077 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32059 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32006 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@32000 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31999 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31998 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31924 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31917 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31916 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31915 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31914 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31906 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31905 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31891 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
significantly improves ethernet<->wireless performance with mac80211 based drivers
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31886 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31856 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31848 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
Should fix #11513. Tested with the x86/generic squashfs image within a VM.
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31841 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
overriding eeprom, fixes bogus wlan mac address on arv4520
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31835 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31813 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31808 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31774 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
master-2012-05-16-2
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31772 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
Dave Täht)
Codel is a new AQM algorithm and RED replacement designed by
Kathie Nichols and Van Jacobson, and published in ACM queue:
http://queue.acm.org/detail.cfm?id=2209336
Codel stands for "Controlled Delay", and needs no knobs in the
general case, twiddled, for optimum results. It aims for 5ms of
delay, at most, when in use.
Additionally,
fq_codel (by eric dumazet) builds on codel to provide fair queuing
superior to what could be had with SFQ, and drop behavior saner
than RED, BLUE, or choke.
These patches are backported from net-next and are known to work
on Linux 3.3.4 and later.
Includes updates to codel for better portability and speed
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31756 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31748 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
It was missing in the previous commit.
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31738 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
* CONFIG_ARCH_DMA_ADDR_T_64BIT
* CONFIG_ARCH_HAS_ILOG2_U32
* CONFIG_ARCH_HAS_ILOG2_U64
* CONFIG_ARCH_PHYS_ADDR_T_64BIT
* CONFIG_ARCH_SUPPORTS_MSI
* CONFIG_GENERIC_CPU_DEVICES
* CONFIG_HWMON_DEBUG_CHIP
* CONFIG_I2C_PXA_PCI
* CONFIG_LEDS_GPIO
* CONFIG_MINIX_FS_NATIVE_ENDIAN
* CONFIG_MLX4_CORE
* CONFIG_POWER_SUPPLY
* CONFIG_POWER_SUPPLY_DEBUG
* CONFIG_QUOTACTL
* CONFIG_SDIO_UART
* CONFIG_SENSORS_K10TEMP
* CONFIG_SENSORS_SCH56XX_COMMON
* CONFIG_SENSORS_VIA_CPUTEMP
* CONFIG_SERIO_AMBAKMI
* CONFIG_SERIO_I8042
* CONFIG_SERIO_LIBPS2
* CONFIG_SERIO_PCIPS2
* CONFIG_SERIO_RAW
* CONFIG_SERIO_SERPORT
* CONFIG_SPI_BITBANG
* CONFIG_SPI_GPIO
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31737 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
It is used by the broken omap35xx/gumstix target only.
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31717 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31713 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31710 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31709 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
The 0x4322 ID is found on the BCM4322 used on some SoCs like the
Linksys WRT610N V1 connected to a BCM4705. The 43222 (0xa8d6) ID
is found on the BCM43222 used on some other SoCs like the D-Link
DSL-2760U connected to a BCM63xx.
This fixes #10887.
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31695 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
It is used only by the broken coldfire target.
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31688 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31687 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31654 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31645 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
having to keep non-upstream ethernet driver changes
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31637 3c298f89-4303-0410-b956-a3cf2f4a3e73
|
|
newer kernels
git-svn-id: svn://svn.openwrt.org/openwrt/trunk@31625 3c298f89-4303-0410-b956-a3cf2f4a3e73
|