This replaces old bcm53xx patch for scanning whole flash and makes
bcm47xxpart compatible with NAND.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47800
Starting with r47755 the pppoe wan interface section isn't any longer added to
the network config on firstboot, using the default configuration. Since the
commit message does not mention such a change, I assume that this wasn't
intentionally.
Signed-off-by: Mathias Kresin <openwrt@kresin.me>
SVN-Revision: 47793
Now that we have redistributable vdsl/adsl firmware blobs in /lib/firmware,
we can drop the dsl_fw partition and extend the firmware partition.
Signed-off-by: Andre Heider <a.heider@gmail.com>
SVN-Revision: 47783
Now that IPV6 isn't build as a module anymore, we don't need to decouple
bridge from ipv6 anymore.
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 47751
In old days mtd_add_partition was checking for partitions overlapping
which was breaking our splitting feature. We had to modify this function
by adding an extra bool dup_check parameter. Upstream commit:
3a434f66e6da ("mtd: part: Remove partition overlap checks")
removed that check so we don't need our modification anymore.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47748
For a long time we have a dedicated firmware splitter for uimage as a
replacement for the one builtin into mtd subsystem. For a year we got a
proper warning for users sticking to the old one (r43700). It should be
really safe to drop it now, especially since we do it for 4.4 only.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47747
Include /lib/functions/system.sh in 02_network to define missing
mtd_get_mac_binary() procedure.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47723
This removes calls to ucidef_set_interfaces_lan_wan() and
ucidef_set_interfaces_lan() on boards where all relevant info can be
inferred from the switch definition.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47722
This changes uci-defaults-new.sh, config_generate and all relevant board.d
files in order combine ucidef_add_switch() and ucidef_add_switch_ports() into
a single function.
Also removes now superfluous enable and reset arguments.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47721
Out of 69 switch definitions, only 3 pass something different than "1" as
values for reset and enable, with one of those three being invalid.
This change ...
* removes the reset and enable arguments from ucidef_add_switch()
* unconditionally writes reset:1 and enable:1 to JSON
* converts the three users of nonstandard values to ucidef_add_switch_attr()
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47720
This migrates the final board.d user of ucidef_add_switch_vlan() to the new
ucidef_add_switch_ports() approach, fixing some logic errors for a few boards
along the way.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47717
Overlayfs stacking was broken in 4.1.12 but now fixed in upstream branch.
Stacking allows experimenting with configurations without the need to
reset a device to factory config.
Signed-off-by: Roman Yeryomin <roman@advem.lv>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47714
This patch adds support for the NuCom R5010UNv2.
It's a BCM6328 based board. It has an onboard BCM43217 wifi chip. For this
wifi chip looks like the brcmsmac driver isn't still supported, b43 drivers are used
for the profile of the router.
It's worth mentioning this board was affected by a bug solved with
https://dev.openwrt.org/changeset/46707
Tested-by: Angel Fontan <angel.fontan@gmail.com>
Signed-off-by: Daniel Gonzalez <dgcbueu@gmail.com>
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 47697
Eth0 is attached to mac0 of the switch with a fixed link and and not to
phy4 in single phy mode, so configuring it to anything but 1000FD will
break the connection, which will happen if a only 100 Mbit capapble device
is plugged into the wan port.
Fix this by not taking the state from phy4 and just configuring a fixed link
for eth0.
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 47695
This updates the iProc PCIe driver to the version currently submitted
for kernel 4.5.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47688
Broadcom submitted new SMP patches for this SoC to upstream Linux, add
them to OpenWrt.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47687
This also deactivates some workaround for erratas only seen in older
CPU cores and L2 cores not used in this SoC.
This partly reverts commit r44947.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47686
This patch is currently placed directly behind the SMP patches and has
nothing to do with SMP, move it to an other position.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47685
netis WF-2881 is an MT7621AT based router with MT7602EN, MT7612EN.
It has 128MB DDR3, 128MB SLC NAND FLASH, 5-port Gbps switch and 1x USB 3.0.
The following patch adds support for this device.
this device only works on top of UBI.
Tested and working:
* ethernet
* both WiFi radios
* USB 3.0
* buttons (reset button)
* ethernet switch and USB diag LEDs
* UART
* GPIOs
* sysupgrade
Tested and not working
* failsafe
Signed-off-by: YounJae Rho <luxflow@live.com>
SVN-Revision: 47619
V2:
The KD-20 also has USB 3.0.I modiy the set_irq_affinity usb 2 to set_irq_affinity ?hci_hcd 2 to bind usb 2.0&3.0 to cpu1.
Signed-off-by: Shonn Lu <countrysideboy@qq.com>
SVN-Revision: 47618
Changes since V1:
I resend this patch for current trunk.The former V1 patch is for CC branch.
Signed-off-by: Shonn Lu <countrysideboy@qq.com>
SVN-Revision: 47617
Roll back the mvebu conversion to board detection code as we need to fix
some strcutural issues first.
Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 47597
This patch introduces support of new boards with ARC cores.
[1] Synopsys SDP board
This is a new-generation development board from Synopsys that
consists of base-board and CPU tile-board (which might have a real
ASIC or FPGA with CPU image).
It sports a lot of DesignWare peripherals like GMAC, USB, SPI, I2C
etc and is intended to be used for early development of ARC-based
products.
[2] nSIM
This is a virtual board implemented in Synopsys proprietary
software simulator (even though available for free for open source
community). This board has only serial port as a peripheral and so
it is meant to be used for runtime testing which is especially
useful during bring-up of new tools and platforms.
What's also important ARC cores are very configurable so there're
many variations of options like cache sizes, their line lengths,
additional hardware blocks like multipliers, dividers etc. And this
board could be used to make sure built software still runs on
different HW configurations.
Cc: Felix Fietkau <nbd@openwrt.org>
Cc: Jo-Philipp Wich <jow@openwrt.org>
Cc: Jonas Gorski <jogo@openwrt.org>
Signed-off-by: Alexey Brodkin <abrodkin@synopsys.com>
SVN-Revision: 47589
Support is present in 3.19 and above as kernel commit:
fcc87a95195236b0935183361a72e4a98bf577d8
Changes to other existing patches are a result of "make target/linux/refresh".
Signed-off-by: Srdjan Rosic <srdjan.rosic@gmail.com>
Signed-off-by: Srdjan Rosic <rosic@google.com>
SVN-Revision: 47588
* use kernel buildin transfer_one_message. we only need to implement
transfer_one and set_cs function
* should support use gpio as cs pin
* deselected the spi device when setup and add debug info
* only reset device when first driver probe
Signed-off-by: Michael Lee <igvtee@gmail.com>
SVN-Revision: 47579
before spi transfer. use spi_prepare_message to setup spi hardware. it
will setup MSB, spi mode and speed
remove sys_freq member and speed check code
Signed-off-by: Michael Lee <igvtee@gmail.com>
SVN-Revision: 47578
* check clock rate, SPI mode, and word sizes
* setup spi polarity
* enable spi1 hw if need
Signed-off-by: Michael Lee <igvtee@gmail.com>
SVN-Revision: 47577
* fill struct according to the member order
* add error clean up
* set min/max spi speed. so we don't need to check again
Signed-off-by: Michael Lee <igvtee@gmail.com>
SVN-Revision: 47576
use loops_per_jiffy, spi clock speed and write bytes to get the
spi loop count. if loop to 0 than spi operation timeout.
remove usleep. we only write 1 byte to spi device. use busy loop
would be better.
Signed-off-by: Michael Lee <igvtee@gmail.com>
SVN-Revision: 47575
This patch adds the posibility to config the led behaviour of the lantiq phys
over the devicetree bindings.It patches the 0023-NET-PHY-adds-driver-for-lantiq-PHY11G
for 3.18 and 4.1. I have also added a register description overview for
applicable values. This goes to the documentation folder in the linux kernel.
Signed-off-by: Florian Eckert <Eckert.Florian@googlemail.com>
SVN-Revision: 47566
It turned out that mtk_nand driver uses some sophisticated accounting and a
general nand code must be patched.
This patch adds required read and erase calls to a general nand code. I
used a code for re6500 released by Linksys as a reference.
All required operations (erase, write and read) are usable. However I found
that jffs2 filesystem can be created only on top of a ubi volume. I tried
to create jffs2 directly on mtd device but pages with clean markers are
became uncorrectable.
Signed-off-by: Kirill Berezin <fyiwdt@gmail.com>
SVN-Revision: 47565
This patch improves the default configuration of DWC2 on lantiq SoCs
somewhat:
* Set maximum packet count to largest allowed value by the DWC2 (511)
* Use 16-bit DMA bursts
* Divide fifo buffers more evenly
Default fifo buffer sizes from original ltq-hcd driver seem really
irrational. For example according to DWC2 data book rxfifo size of 240
will not fit even a single full length USB packet. On the other hand
non-periodic tx fifo size of 240 is more than enough to fit one complete
packet.
Change the sizes around to improve the situation and to fix some issues
especially with isochronous USB transfers.
Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
SVN-Revision: 47563
The hwid check was wrong, causing the AR71XX_MODEL value to end with a
space (as $hwver was unset).
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
SVN-Revision: 47553
r47218 was missing needed changes which were accidentally
ommitted from the patch.
This should fix#20878
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 47552
When fixing the model string for WNDR3700v2 which contain a model string
followed by 0xff in r46455, the match for other versions of the WNDR3700v2
which just contain lots of 0xff broke (as the 0xff $model is checked for
is stripped off).
Fix by stripping off non-printable characters only for the actual output
string, but not for the internal matching.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
SVN-Revision: 47538
Removed video related kernel params that are only appropriate for the
Freescale downstream vendor kernel and not appriate for mainline linux.
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
SVN-Revision: 47532
The 206-ARM-imx-ventana-added-GW16083-to-device-tree.patch merged wrong in
the conversion from 3.18 to 4.1. This patch resolves that and puts the
device-tree nodes in the correct place.
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
SVN-Revision: 47510
Replace the ventana ubi images from a single rootfs volume containing the
rootfs as well as kernel+dtbs in the boot/ subdir to a multi-volume ubi
containing kernel+dtbs+bootscript in 'root' ubifs volume, rootfs in 'rootfs'
squasfs volume, and rootfs_data an empty ubifs volume.
At the same time move ventana image creation over to using the new
TARGET_DEVICE syntax as a cleanup.
Additionally removed FIT image creation as it is recommended that this be
done manually to produce signed FIT images if needed.
Note that a bootscript is needed for Ventana to change its default mtdparts
to rename the 2nd partition from 'rootfs' to 'ubi'. This is because there
are patches in OpenWrt kernels that try to auto-attach and auto-mount and these
get in the way if the mtd partition is named 'rootfs'.
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
SVN-Revision: 47508
Upstream patch: e9d6d6b62f306ba83e1441af5daf2809a6167474
Add support for the Gateworks GW5510 board featuring:
* i.MX6 SoC
* up to 512MB DDR3
* up to 2GB NAND flash
* 1x miniPCIe socket (with USB)
* HDMI out (micro-HDMI)
* HDMI in (micro-HDMI) (currently supported by only vendor kernel)
* TTL level I/O (supported by GW16111 breakout board):
* I2C
* 2x UART
* CAN
* 2x DIO (GPIO/PWM)
* USB OTG
Also add support to OpenWrt build system
Signed-off-by: Tim Harvey <tharvey@gateworks.com>
Signed-off-by: Pushpal Sidhu <psidhu@gateworks.com>
SVN-Revision: 47505
compat-wireless/backports now contains a bcm47xx_nvram.h file to
backport some of the functions in it which are used by the bcmfmac
driver. This file just checks for the kernel versions and provide an
empty implementations on older kernel versions. This is OK on most
systems, but on bcm47xx / bcm53xx systems we want to call the real
functions here. This commit removes the file from backports in our
build process like we do it with the bcma and ssb header files. Instead
we add a recent version into our kernel so all code uses only one
header file. On bcm47xx / bcm53xx the real implementations of this code
will be used.
Reported-by: Hante Meuleman <meuleman@broadcom.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47467
This is based on patches from Federico Fissore <f.fissore@arduino.cc>
especially this one:
7e2976fa83
The console is running with 250000 baud which is a non standard baud
rate and needs an extra patch to be applied, I will try to get this
patch upstream or something else which accomplish the same.
Some upstream code looks like there are many different versions of this
SoC, are these only internal versions and all versions on the consumer
market are the same? I saw different GPIO configuration and flash sizes
of 8MB and 16MB?
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47451
The Arduino Yun uses a baud rate of 250000 by default. The serial is
going over the Atmel ATmega and is used to connect to this chip.
Without this patch Linux wants to switch the console to 230400 Baud.
With this patch Linux will use the configured baud rate and not some
standard one which is near by.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47450
Add access to the function2 gpio register. This probably has to be
converted into a pimux driver later on. This is needed for some setup
functions on the Arduino Yun.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47449
This automatically writes to the alternate firmware partition
(on Linksys's dual firmware layout) using code borrowed from
the WRT1900AC port.
Signed-off-by: Claudio Leite <leitec@staticky.com>
SVN-Revision: 47434
This is done with existing code from the WRT1900AC port.
It makes sure the "auto_recovery" bootloader option is set,
and resets the s_env boot counter after a successful boot.
This gives users without a serial console connection some
measure of safety.
Signed-off-by: Claudio Leite <leitec@staticky.com>
SVN-Revision: 47433
This is done with the new image generation system and board names.
It also generates a sysupgrade.tar image.
Signed-off-by: Claudio Leite <leitec@staticky.com>
SVN-Revision: 47432
This is imported verbatim from the mvebu WRT1900AC port.
It picks up the current boot partition from the kernel command line,
then renames that partition "ubi" so that it auto-mounts.
Signed-off-by: Claudio Leite <leitec@staticky.com>
SVN-Revision: 47431
- Use board engineering names rather than marketing names
- Linksys uses a dual firmware layout, where the bootloader
will switch to the other stored image when one fails to
boot three consecutive times.
In order to make this firmware compatible with the factory
images and the stock bootloader we must match this layout.
Signed-off-by: Claudio Leite <leitec@staticky.com>
SVN-Revision: 47429
The international version is completely different from the already
supported Chinese version. The WLAN of the QCA956x SoC used by this router
has been fixed in r46948.
This patch looks like it changes a lot in
700-MIPS-ath79-openwrt-machines.patch; that is not the case. Unfortunately,
quilt decided to completely reorganize the Kconfig patch even though only
a single section has been added.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
SVN-Revision: 47420
* remove references to non-existing kmod-usb3-mt7621
* ZTB WG2626 doesn't have an RTC (thanks to Piotr Dymacz!)
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 47407
The hardware of the v3 is identical to the v2.
Based-on-patch-by: Andreas Ziegler <github@andreas-ziegler.de>
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
SVN-Revision: 47406
Some devices require the kernel to be in a JFFS2 file system. Make the
support for this more generic so that it can be used by other devices.
Signed-off-by: Benjamin Berg <benjamin@sipsolutions.net>
SVN-Revision: 47404
the board names used by the upgrade check differ from the device names
in case, so always set them. We can't rely on tolower or similar since
different locales might have different rules for lower case letters, so
set them explicitily.
Fixes sysupgrade rejecting *-sysupgrade.tar because of the difference.
Signed-off-by: Kaspar Schleiser <kaspar@schleiser.de>
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 47381
Incorrect value causes clock inaccuracy as huge as 1/60.
Signed-off-by: Dmitry Ivanov <dima@ubnt.com>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 47363
the support is still WIP. next steps are to make the pmic and ethernet work.
this is the first commit to make sure nothing gets lost.
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 47354
This patch adds the eeprom property to the dts for the Asus RP-N53. It is necessary to get the wifi in the soc working.
Signed-off-by: Alberto Mattea <alberto@mattea.info>
SVN-Revision: 47350
Replace use of vconfig with the corresponding iproute2 syntax in
/lib/preinit/07_set_preinit_iface_ramips
The switch reset work-around needed on some ramips targets used the
deprecated vconfig tool to bring up the pre-init/failsafe interface.
This doesn't work on recent builds where vconfig is no longer being
deployed, resulting in the device being unreachable when triggered
into failsafe mode.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 47349
Add support for Dovado tiny AC wifi router.
Soc: mt7620a
ram: 64MB
flash: 8MB
1x usb 2.0
two gigabit lan ports
5ghz wlan is not supported on this board since there is no gpl driverfor mt7610e wifi chip.
Signed-off-by: Andrej Vlasic <andrej.vlasic0@gmail.com>
SVN-Revision: 47348
Some boards like dovado tiny ac have generic ralink mac in onflash wifi eeprom.
Add eeprom_set_macaddr function to rt2x00 eeprom extract script which changes mac addr insde eeprom dump.
Signed-off-by: Andrej Vlasic <andrej.vlasic0@gmail.com>
SVN-Revision: 47347
Add missing icplus phy driver kernel config symbol to mt7620 subtarget.
This is needed since some mt7620 devices have icplus ip1001 ethernet transceiver
Signed-off-by: Andrej Vlasic <andrej.vlasic0@gmail.com>
SVN-Revision: 47346
This patch adds a case for the Asus RP-N53 in the "02_network" boot script. Without this, the lan interface does not get configured on startup, effectively bricking the device.
Signed-off-by: Alberto Mattea <alberto@mattea.info>
SVN-Revision: 47345
The device is similar to Archer C7 v2, but with 8M of flash, and 6 external
antennas. Reason for breaking the naming pattern is that sticker on the back
mentions "WDR7500" and has no mention of "Archer".
Signed-off-by: Srdjan Rosic <rosic@google.com>
SVN-Revision: 47344
This fix contains 2 parts:
- kernel 4.1: backport upstream patch "perf build: Do not fail on missing Build file"
- add NO_LZMA=1 to perf MAKE_FLAGS to disable LZMA support
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
SVN-Revision: 47338
Many packages for OpenWrt currently assume that lan is the "internal"
side of the network map, e.g. if installed, dnsmasq will hand out DHCP
leases on lan, firewall will allow forwarding on lan.
While at it, also configure a lan interface if eth1 exists.
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
SVN-Revision: 47326
There is currently no relevant definition for developerboard device type
in include/target.mk, thus packages like iptables, firewall, dnsmasq
etc. are missing in the default images, rendering it almost useless
for regular testing or normal use.
With this change, currently a default set of packages for "router" will
be installed to malta firmware images.
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
SVN-Revision: 47325
There are instructions (e.g. wsbh) only available in MIPS32R2 which is
the isa level used by many MIPS targets in OpenWrt. As we use the malta
target mostly for development and testing purposes, setting the CPU_TYPE
to 24Kc will make the emulation more real.
The following is output from qemu-system-mipsel 2.3.0
root@OpenWrt:/# cat /proc/cpuinfo | grep 'cpu model\|isa'
cpu model : MIPS 24Kc V0.0 FPU V0.0
isa : mips1 mips2 mips32r1 mips32r2
Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
SVN-Revision: 47324
All other tplink images use squashfs+jffs2. I don't see any reason why
this board should use squashfs-only
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
SVN-Revision: 47319
It was wrong anyway, missing file sort order parameter which was fixed
in base month ago.
Signed-off-by: Petr Štetiar <ynezz@true.cz>
SVN-Revision: 47314
It was wrong anyway, missing file sort order parameter which was fixed
in base month ago.
Signed-off-by: Petr Štetiar <ynezz@true.cz>
SVN-Revision: 47313
It was wrong anyway, missing file sort order parameter which was fixed
in base month ago.
Signed-off-by: Petr Štetiar <ynezz@true.cz>
SVN-Revision: 47312
This enables booting from devices that use an ATI PATA controller for
the boot device, such as the embedded CF cards in Fujitsu-Siemens Futro
thin-clients.
Signed-off-by: Felix Kaechele <heffer@fedoraproject.org>
SVN-Revision: 47304
Following the additions in r46924, enable the 'ext4' feature for
realview target so the ext4 image is created by default.
Signed-off-by: Dov Murik <dov.murik@gmail.com>
SVN-Revision: 47301
According to the info from NVRAM we should use port 8 for the CPU (and
interface eth2). Unfortunately it doesn't work right now, so lets switch
to the port 5.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Forwardport of r46586 from 15.05
SVN-Revision: 47281
Kernel 3.14 added aditional genphy_soft_reset phy reset to phy_init_hw in drivers/net/phy/phy_device.c
Since adm6996 does in driver soft reset and doesn't use BMCR_RESET for soft reset
add dummy soft_reset callback to adm6996 driver, like it is done in ar8216.
This fixes ticket #20147
Signed-off-by: Andrej Vlasic <andrej.vlasic0@gmail.com>
SVN-Revision: 47272
Observed on ar71xx/ath79 platforms such as Ubiquiti RouterStations.Reported in #20642.:
(https://dev.openwrt.org/ticket/20642).
If embedded command-line text exists with CONFIG_IMAGE_CMDLINE_HACK=y,firmware
init doesn't initialize environment pointer (fw_init_cmdline not called).
arcs_cmdline is not initialized before calling strlcat.
Signed-off-by: Ted Hess <thess@kitschensync.net>
Tested-by: Stijn Tintel <stijn@linux-ipv6.be>
SVN-Revision: 47270
The TL-WR841N/ND v10 is mostly identical to the v9. Apart from some minor
changes, it contains a newer revision of the QCA9533 SoC and the CPU clock
is significantly higher.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
SVN-Revision: 47263
ath9k expects to get revision id 2 for the QCA9531 ver. 2 rev. 0. This
fixes the very low TX power on some devices like the TP-LINK
TL-WR841ND v10.
As ath79_soc_rev is only used to get the revision number to ath9k on the
QCA9533, just set it to the expected value on the ver. 2.
Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
Tested-by: Felix Kaechele <felix@kaechele.ca>
SVN-Revision: 47262
Antminers using the stock bootloader will not hash without this GPIO set.
Applies to DD and CC
Signed-off-by: L. D. Pinney <ldpinney@gmail.com>
SVN-Revision: 47261
This adds a basic implementation of clk_round_rate()
The clk_round_rate() function is called by multiple drivers and
subsystems now and the lantiq clk driver is supposed to export this,
but doesn't do so, this causes linking problems like this one:
ERROR: "clk_round_rate" [drivers/media/v4l2-core/videodev.ko] undefined!
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47254
These patches are written by Broadcom and will be in mainline Linux
kernel soon. I had some problems to get them backported to kernel 4.1,
so currently they are only available for 4.3.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47253
The files directory is now split up into the files which are needed for
every kernel version and the files only needed by kernel 4.1. The files
in files-4.1 are already merged into mainline kernel 4.3. This patch
only removed patches which were merged into mainline kernel 4.3.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47251
This replaces our SMP implementation with a very similar version which
is Broadcom currently trying to get into mainline kernel.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47247
Add device tree files for Broadcom Northstar based SVKs. Since the
bcm5301x.dtsi already exists, all that is necessary is the dts files to
enable the UARTs. With these files, the SVKs are able to boot to shell.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 47246
This patch fixes ticket #15267 by enabling power on the
WNR2200's USB port. At present, the USB port on the WNR2200
is non-functional due to it not receiving power.
This patch defines an additional GPIO pin, but none of the
current GPIO definitions have been modified.
Signed-off-by: Riley Baird <BM-2cVqnDuYbAU5do2DfJTrN7ZbAJ246S4XiX@bitmessage.ch>
SVN-Revision: 47236
This patch supersedes the v1 from September 17th.
Bumping the patch version - the MiniBox profile showed up under M, but
since it's called 'Gainstrong MiniBox v1.0' now it looks out of place.
Renamed the profile to gs-minibox-v1.mk to fix that.
The following patch adds support for the Gainstrong MiniBox into trunk
(or 'Designated Driver' :D ).
Fixed items:
- Inverted LED polarity (OOLITE seems to suffer from the same problem).
- Changed uppercase MINIBOX_V1_ prefix as requested.
- Prefixes are now gs_minibox_ similar to gs_oolite_ (same vendor).
- Mention the vendor (Gainstrong) in code headers.
Compiles fine, has been confirmed working by owners on 15.05.
Question: I've seen some boards use tools/firmware-utils/src/mktplinkfw.c,
the MiniBox images build fine without, so I'm wondering: do I need to add
it in there as well? Any added benefit?
Thank you
Signed-off by: Stijn Segers <francesco.borromini@inventati.org>
SVN-Revision: 47234
Sets the LEDs to boardname:color:led-name
Sets the LAN to eth0
Other corrections such as the Machine Name and HWIDs
v2 corrects the profile names in the Makefile and changes tabs to spaces
in the Makefile 'define Device/' like the other devices.
Signed-off-by: L. D. Pinney <ldpinney@gmail.com>
SVN-Revision: 47221
This contains two sets of fixes for the 8139cp driver.
For all kernel versions older than 4.3, we can apply the fixes from the
4.3-rc4 kernel. In particular, these fix the TX timeout recovery which
is causing my Geos to lock up until the hardware watchdog kicks in.
For 4.0 and later kernels, we can also apply the additional improvements
which are going into 4.4 to fix and enable hardware checksum/TSO
offload. Backporting those to older kernels is non-trivial.
Signed-off-by: David Woodhouse <David.Woodhouse@intel.com>
SVN-Revision: 47220
Still a lot of kernel-version ifdef'ery, but imho that's easy to remove
once obsoleted and avoids duplicate code in the meantime.
Signed-off-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 47218
Whenever two processes were executing different commands at the same
time then one of the commands sometimes got the response of the other
command.
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
SVN-Revision: 47213
This should help finding potential problems with the SPI driver.
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
SVN-Revision: 47209
Until now the SPI driver used the TX bits for the RX FIFO. This seems
uncritical for now since both are equals on my devices (VR9), but this
could cause problems on other SoCs.
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
SVN-Revision: 47208
This reverts commit 68c2e4789b4f071ee75d39248f4d08fe8283eb28.
commit r47159 was bad
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 47207
add the SDK alsa driver. this has only been tested on mt7628/88 and wm8960.
mt7620 is only compile tested.
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 47205
When talking to an atmel controller we need 9600 or 250000 baud.
as 250000 does not exist we use 2500000.
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 47204
* use build step mktplinkfw with extra arguments
* remove now unused mktplinkfw-chn-v2
* use Device/tplink
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
SVN-Revision: 47175
Allow to pass extra arguments to mktplinkfw step. Some board requires
an extra argument to create a valid image.
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
SVN-Revision: 47174
These profiles covered by new image/Makefile descriptions. Only the old format uses
MultiProfile macro
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
SVN-Revision: 47173
This patch fixes a wrong non pre-emptive crc errors output of
dsl_control.sh status.
Signed-off-by: Luca Debernardi <luca.debernardi@gmail.com>
SVN-Revision: 47172
The device is similar to the TD-W8970, beside a different Atheros 2.4 GHz
wireless chip and the additional, PCI connected, WAVE300 5 GHz wireless.
Signed-off-by: Mathias Kresin <openwrt@kresin.me>
SVN-Revision: 47130
Port of r41856.
In contrast to the brcm63xx target, it isn't sufficient to enable/disable
the bridge. The device needs to be enabled/disabled to fix the hang. The
bridge will be automatically enabled by the time the connected device is
enabled.
Fixes boot on TD-W8980.
Signed-off-by: Mathias Kresin <openwrt@kresin.me>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
SVN-Revision: 47129
This adds the changes from r46219 to the linux 4.1 patches as well.
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
SVN-Revision: 47128
This way it's easier to configure device tree overlays, customize other
parameters...
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 47126
Right now, selecting kmod-sound-soc-bcm2708-i2s causes build errors due to
missing configs.
kmod-regmap enables I2C and SPI, causing build errors due to depending
variables not defined.
Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 47123
when sleep mode is disable use MIPS as clocksource and clockevent instead of systick.
because MIPS timer has higher resolution 5ns less than systick 20us and
larger counter bits 32 > 16.
clean interrupt by write compare register at isr.
fix typo cause sleep mode not enable.
Signed-off-by: Michael Lee <igvtee@gmail.com>
SVN-Revision: 47122
Due to datasheet of rt3883 SoC rgmii1 port handles pins 84-95 and rgmii2 port handles pins 72-83. When this function ports gets added to rt3883_pinmux_data there's wrong pinmux bits set (RT3883_GPIO_MODE_GE1 manages 84-95 pins and RT3883_GPIO_MODE_GE2 manages 72-83). So when enabling rgmii2 as GPIO driver confuses hardware and nothing work, neither rgmii nor gpio.
Also in '0030-pinctrl-ralink-add-pinctrl-driver.patch' typo in name of rgmii2 port.
Signed-off-by: Nick Leiten <nickleiten@gmail.com>
SVN-Revision: 47118
The default switch frame size (with FCS + header) is 1536 bytes. But the
GMAC only accepted frames up to 1522 bytes. Setting it to 1536 allows to
receive ethernet frames using the full of MTU 1500 + an extra VLAN header +
VLAN header added by the switch.
Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 47117
The driver assumes that the maximum received buffer for non-jumbo frames is
1536 bytes. But the allocation of the rx fragment doesn't reflect that. It
currently allocates fragments which will only be large enough to be used as
rx buffer with the size of 1534 bytes. This is problematic because the GMAC
will now try to write to 2 bytes which don't belong to its receive buffer
when a large enough ethernet frame is received.
This may already be a problem on existing chips but will at least become a
problem when the 1536 byte rx modus is enabled on MT7621a. It is required
on this SoC to receive ethernet frames which use their full 1500 bytes MTU
and a VLAN header next to the switch VLAN tag.
Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 47116
The length of the DMA rx buffer was always set to 0 because the function
for extracting the length was used to calculate the value for setting it.
Instead the macro has to be split in a get and set function similar to the
TX_DMA_(GET_|)PLEN(0|1) macro.
No problem was noticed on MT7621a before this was changed and thus maybe it
was hidden by different problem which is not yet fixed.
Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 47115
The MT7530 switch driver with enable_vlan set will automatically set all
ports to the user port mode. The hardware will remove the incoming vlan tag
on these ports and use it for its internal vlan. This is usually not wanted
and makes it impossible to communicate via vlan over the switch in both
directions.
It is possible to configure a switch port to "transparent mode" when this
port is only used as untag in the switch VLANs. This will disable the VLAN
untagging of packets when they were received on this port. The tagging on
"tag" ports based on the vlan id is still working.
The transparent port mode cannot be used when a port is both used in a VLAN
as "tag" and in another one as "untag" port.
Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 47114
HiWiFi HC5661/5761/5861 models are manufactured by http://www.hiwifi.com <http://www.hiwifi.com/>. These models have similar hardware specs(MT7620A + 128M DDR2 + 16M flash). This patch adds support for them.
The original author is Justin Liu (rssnsj@gmail.com). I ported the patch to trunk and submitted it here with his approval.
v3 fix
1: Spaces -> Tabs
2: Removed some packages
Signed-off-by: Xiaoning Kang <kangxn@163.com>
SVN-Revision: 47113
HiWiFi HC5661/5761/5861 models are manufactured by http://www.hiwifi.com. These models have similar hardware specs(MT7620A + 128M DDR2 + 16M flash). This patch adds support for them.
The original author is Justin Liu (rssnsj@gmail.com). I ported the patch to trunk and submitted it here with his approval.
v3 fix
1: Merged most stuff into dtsi file
2: Remove unnecessary empty lines.
Signed-off-by: Xiaoning Kang <kangxn@163.com>
SVN-Revision: 47112
HiWiFi HC5661/5761/5861 models are manufactured by http://www.hiwifi.com. These models have similar hardware specs(MT7620A + 128M DDR2 + 16M flash). This patch adds support for them.
The original author is Justin Liu (rssnsj@gmail.com). I ported the patch to trunk and submitted it here with his approval.
v3 fix
1: Fixed model order
2: Remove manufacturer name from model name
3: Use a hacky but prettier way to get mac address.
Signed-off-by: Xiaoning Kang <kangxn@163.com>
SVN-Revision: 47111
Only add 'cuImage.tl-wdr4900-v1' to KERNEL_IMAGES if
the selected subtarget is 'generic'.
Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
SVN-Revision: 47109
A year ago I made this old splitter obsolete (see r42828), it's time to
completely remove it now. For all this time we got these warnings:
"Dedicated partitioner didn't create "rootfs_data" partition, please fill a bug report!"
and
"Support for built-in "rootfs_data" splitter will be removed, please use CONFIG_MTD_SPLIT_SQUASHFS_ROOT"
but I'm not aware of any bug reports related to that.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47040
We're in process of cleaning nand_do_platform_check. Currently it leaves
a special mark in /tmp/sysupgrade-nand-path triggering some diffent code
path in nand_upgrade_stage1. This can be a bit confusing.
The plan is to have the check function only check the image and nothing
else. Then platform code (platform_pre_upgrade) should trigger NAND
specific upgrade path. This is what this patch implements.
This follows all other targets making it finally possible to change
nand_do_platform_check.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 47039
First of all platform_do_upgrade_phase2 doesn't seem to be defined
anywhere. Other than that platform_do_upgrade is not likely to be ever
called. There is a call to nand_do_platform_check which triggers
nand_upgrade_stage1 call with "exit 0" in it.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Acked-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 47038
Add support for WeIO board (http://we-io.net).
This board is based on Carambola2 board form 8Devices.
Signed-off-by: Drasko DRASKOVIC <drasko.draskovic@gmail.com>
Signed-off-by: Felix Fietkau <nbd@openwrt.org> [some cleanups]
SVN-Revision: 47036
- Duckbill uses a different partitioning approach than standard
FSL and Olimex
- use new sdimage to integrate U-Boot into the SD card images
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
SVN-Revision: 47021
- fix Kernel dts subdir location
- enable virtio networking
we are using -M virt which supports virtio devices
changes in symbols due to make kernel_oldconfig:
- CONFIG_ARCH_BINFMT_ELF_RANDOMIZE_PIE=y : removed
fix security: upstream id: d1fd836dcf00d2028c700c7e44d2c23404062c90
- CONFIG_ARM64_CPU_SUSPEND: removed
commit id: af3cfdbf56b91785650f54e7c9a899d814b4b9fb
- CONFIG_ARM64_PGTABLE_LEVELS=3 : renamed PGTABLE_LEVELS
commit id: 9f25e6ad58e1fb3b4d441e4c55635c4598a6fa94
- CONFIG_GENERIC_CLOCKEVENTS_BUILD=y : removed upstream
commit id: 9f083b74df3a7eaa100b456f2dc195512daf728e
-# CONFIG_PREEMPT_RCU: unnecessary
unknown remove reasons:
-CONFIG_GENERIC_IOMAP=y
-# CONFIG_POWER_RESET_GPIO is not set
-# CONFIG_POWER_RESET_GPIO_RESTART is not set
-# CONFIG_POWER_RESET_LTC2952 is not set
-# CONFIG_POWER_RESET_SYSCON is not set
-# CONFIG_XEN is not set
tested and boot, networking work with qemu
Signed-off-by: Dirk Neukirchen <dirkneukirchen@web.de>
SVN-Revision: 47005
QCA9563 and QCA9561 are two series of Qualcomm SoC Dragonfly. The only different
is QCA9563 w/o internal switch. It has one GMAC with SGMII interface. But they
have the same device ID(0x1150). So they share the same codes.
Signed-off-by: Miaoqing Pan <miaoqing@codeaurora.org>
SVN-Revision: 46971
the lantiq dsl hack to struct atm_trafprm broke the ABI fix this by moving
the variables to the end of the struct
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 46964
The pinctrl-rt2880 code doesn't support multiple functions with the same
name. This will result in a incorrect pinmux configuration.
Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 46963
This patch adds kernel and userspace support for the Ubiquiti Networks airGateway Pro.
Signed-off-by: Matthew Reeve <mreeve at tenxnetworks.com>
SVN-Revision: 46961
We're in process of cleaning nand_do_platform_check. Currently is leaves
a special mark in /tmp/sysupgrade-nand-path triggering some diffent code
path in nand_upgrade_stage1. This can be a bit confusing.
The plan is to have the check function only check the image and nothing
else. Then platform code (platform_pre_upgrade) should trigger NAND
specific upgrade path. This is what this patch implements.
This follows ar71xx, lantiq and partially bcm53xx.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46945
We're in process of cleaning nand_do_platform_check. Currently is leaves
a special mark in /tmp/sysupgrade-nand-path triggering some diffent code
path in nand_upgrade_stage1. This can be a bit confusing.
The plan is to have the check function only check the image and nothing
else. Then platform code (platform_pre_upgrade) should trigger NAND
specific upgrade path. This is what this patch implements.
This follows ar71xx, lantiq and partially bcm53xx.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46944
As explained in recent sysupgrade/nand.sh commits, current NAND
sysupgrade is a bit misleading because of nand_do_platform_check
behavior. It leaves a special mark in /tmp/sysupgrade-nand-path
triggering some diffent code path in nand_upgrade_stage1.
The plan is to have the check function only check the image and nothing
else. Then platform code (platform_pre_upgrade) should trigger NAND
specific upgrade path. This is what this patch implements.
Please note that because of current nand_upgrade_stage1 implementation
this patch doesn't change any behaior yet. It only prepares lantiq
target for changing nand_do_platform_check (it will be possible after
preparing all other targets as well).
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46939
This patch adds support for TP-LINK TL-WDR3320 v2.
This router uses a chinese version 2 firmware header,.
Signed-off-by: Weijie Gao <hackpascal@gmail.com>
SVN-Revision: 46934
The GPIO pins for "POE passthrough" of Ubiquiti Nanostation models are the
following:
* Ubiquiti Nanostation M XM: Pin 8
* Ubiquiti Nanostation M XW: Pin 2
The previous definition of the pins was mixed up between XM and XW model.
Signed-off-by: Lars Kruse <lists@sumpfralle.de>
SVN-Revision: 46922
This patch is to add the WIZnet WizFi630A board as a new platform. The board is in mini pci express form factor.
Signed-off-by: Tobias Welz <tw@wiznet.eu>
SVN-Revision: 46921
This patch add support for Planex DB-WRT01. DANBOARD route on
the MT7620A SoC with two Ethernet port and a 802.11n 2.4 GHz radio.
DANBOARD is Cartoon character.
Signed-off-by: YuheiOKAWA <tochiro.srchack@gmail.com>
SVN-Revision: 46918
Add AR9 DTS definition to be recognized by the DWC2 driver.
The same driver parameters can be mostly used except that some boards
seem to erroneously report OTG HNP/SRP capability of the USB HCD.
Forcing the HNP/SRP off allows these boards to work with the DWC2 as well.
Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
SVN-Revision: 46915
Since the AR9 USB is very similar to the VR9 USB it too can be used with
the upstream DWC2 driver.
Here are the DTS definitions which make it compatible with the DWC2
driver.
Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
SVN-Revision: 46914
Based on the ltq-hcd driver the AR9 USB can be initialized the same way
as the VR9 platform. Use the same initialization bits for both
platforms.
Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
SVN-Revision: 46913
To make the lzma-loader build with binutils 2.25.1 it is needed to
remove the .MIPS.abiflags section otherwise I get the following compile
error:
mips-openwrt-linux-musl-objcopy -O binary -R .reginfo -R .note -R .comment -R .mdebug -S loader loader.bin
mips-openwrt-linux-musl-objcopy: Warning: Writing section `.text' to huge (ie negative) file offset 0xffffffff80a00000.
mips-openwrt-linux-musl-objcopy:loader.bin[.text]: File truncated
This should close#20487.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46893
This moves arch patches affecting many DTS files. Thanks to this change
we can put some ready-to-mainline patches at 300-319 and simplify
backporting process.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46889
As requested here #20432 I've
tried to fix the missing symbols from the specified default kernel
config files.
I ran into this problem trying to build an image for x86_64 target
with a custom kernel configuration.
Signed-off-by: Manganini Mirko <mmanganini@gmail.com>
SVN-Revision: 46879
the trx utile uses a maximum image size of 7.2MB. There are brcm47xx
devices even with serial flash with bigger flash chips, but OpenWrt was
not able to create images for these devices. This patch provides an
additional parameter which increases this limit to 32 MB. There is a
warning in the trx utile code which suggests that bigger images could
overwrite the nvram partition on some devices, but normally the program
writing the image should make sure that it is safe to write it to the
flash.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46872
This patch changes nothing on the behaviour, it just breaks long lines
with bin/trx to make it easier to add additional parameters.
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46871
Fixes support for AR9287 on TP-Link TD-W8980 and possibly other devices
which have an ath wifi chip at a PCI address other than 0xb8000000
(TD-W8980 for example has it's wifi chip at 0xbc000000).
Signed-off-by: Geoffrey McRae <geoff@spacevs.com>
Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
SVN-Revision: 46869
The previous "link" and "status" functions were non-standard,
and thus less useful for parsing.
Signed-off-by: Claudio Leite <leitec@staticky.com>
SVN-Revision: 46864
Increase space for kernel and also introduce a 8M flash target into the
Makefile.
Signed-off-by: Attila Lendvai <attila@lendvai.name>
SVN-Revision: 46862
The PBR-M1 and other upcoming MT7621 boards have RTC chips on them. The
PBR-M1 also selects the kmod-rtc-pcf8563 by default. But the module itself
will not be build because CONFIG_RTC_CLASS is currently not enabled for its
kernel.
Enabling this option should fix the problem of the missing rtc device on
these boards.
Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 46857
According to the AR7242 datasheet section 2.8, AR724X CPUs use a 40MHz
input clock as the REF_CLK instead of 5MHz.
The correct CPU PLL calculation procedure is as follows:
CPU_PLL = (DIV * REF_CLK) / REF_DIV / 2.
This patch is compatible with the current calculation procedure with default
DIV and REF_DIV values.
Test on both AR7240, AR7241 and AR7242.
Signed-off-by: Weijie Gao <hackpascal@gmail.com>
SVN-Revision: 46856
This adds full support (sans sysupgrading from vendor firmware) for the COMFAST
CF-E316N v2 (aka CF-E316V2, CF-E316N-V2 and CF-E316Nv2.0, no FCC ID) by
Shenzhen Four Seas Global Link Network Technology Co., Ltd (this company is
actively refusing to provide GPL'd sources for the OpenWrt version they ship
with the device, damn them).
Signed-off-by: Paul Fertser <fercerpav@gmail.com>
SVN-Revision: 46852
In krait_cpufreq_probe, both freq and max_cpu_freq are never
initialized, so the max_cpu_freq will have a random value at the end.
Fix this by properly initializing max_cpu_freq to 0 and storing the clk
frequency in freq as well, to make it similar to how it's calculated in
krait_set_target.
Fixes the following warnings:
In file included from include/linux/clk.h:16:0,
from drivers/cpufreq/cpufreq-krait.c:13:
drivers/cpufreq/cpufreq-krait.c: In function 'krait_cpufreq_probe':
include/linux/kernel.h:714:24: warning: 'freq' may be used uninitialized in this function [-Wmaybe-uninitialized]
_max1 > _max2 ? _max1 : _max2; })
^
drivers/cpufreq/cpufreq-krait.c:217:25: note: 'freq' was declared here
unsigned long freq_Hz, freq, max_cpu_freq;
^
In file included from include/linux/clk.h:16:0,
from drivers/cpufreq/cpufreq-krait.c:13:
include/linux/kernel.h:714:24: warning: 'max_cpu_freq' may be used uninitialized in this function [-Wmaybe-uninitialized]
_max1 > _max2 ? _max1 : _max2; })
^
drivers/cpufreq/cpufreq-krait.c:217:31: note: 'max_cpu_freq' was declared here
unsigned long freq_Hz, freq, max_cpu_freq;
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46839
The phy driver has its qcom-dwc3 order switched in contrast to the usb
controller driver.
Signed-off-by: Kaspar Schleiser <kaspar@schleiser.de>
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46838
commit r30917 ("kernel: bypass all netfilter hooks if the sysctls for that
functionality have been disabled - eliminates the overhead of enabling
CONFIG_BRIDGE_NETFILTER in the kernel config") introduced an optimization
which should reduce/eliminate the overhead for traffic send over bridges on
kernels compiled with CONFIG_BRIDGE_NETFILTER=y. But this optimization
breaks the nf_call_iptables per bridge setting which is more fine grained
than the global sysctl net.bridge.bridge-nf-call-iptables setting.
A test reflecting a real world setup was created to identify if this really
eliminates the overhead and if per-bridge nf_call_iptables could be used in
some setups to increase the throughput. A Qualcomm Atheros QCA9558 based
system with one ethernet and an ath9k wifi 3x3 in HT40 mode was used.
Cables from the AP to the wifi station were used to reduce interference
problems during the tests.
The wlan interface was put in one bridge interface called br-wlan. This
bridge usually contains some more wlan interfaces. The eth0 was put in a
second bridge called br-lan. This usually contains some other privileged
wlan or mesh interfaces. Routing was added between br-lan and br-wlan.
Three kernels were tested:
* (default) OpenWrt kernel for this device
* (brfilter-global) OpenWrt kernel with CONFIG_BRIDGE_NETFILTER=y
* (brfilter-local) OpenWrt kernel with CONFIG_BRIDGE_NETFILTER=y and
without 644-bridge_optimize_netfilter_hooks.patch
The changes to the the netfilter settings of the bridge were done via:
* (brfilter-global) /sbin/sysctl -w net.bridge.bridge-nf-call-iptables=1
* (brfilter-lobal) echo 1 > /sys/class/net/br-lan/bridge/nf_call_iptables
and/or echo 1 > /sys/class/net/br-wan/bridge/nf_call_iptables
A station connected to the wlan0 (AP) interface was used to send traffic to
a PC connected via ethernet. iperf with 3 concurrent transmissions was used
to generate the traffic.
| kernel | br-nf-* global | nf-call* iface | download | upload |
|-----------------|----------------|----------------|----------|----------|
| default | 0 | - | 209 | 268 |
| brfilter-global | 0 | - | 185 | 243 |
| brfilter-local | 0 | - | 187 | 243 |
| brfilter-local | 0 | br-lan | 157 | 226 |
| brfilter-local | 0 | br-lan br-wlan | 139 | 161 |
| brfilter-global | 1 | - | 136 | 162 |
Download/upload results in Mibit/s
It can be seen that the patch doesn't eliminate the overhead. It can also
be seen that the throughput of brfilter-global and brfilter-local with
disabled filtering is the roughly the same. Also the throughput for
brfilter-global and brfilter-local for enabled filtering on all bridges is
roughly the same.
But also the brfilter-local throughput is higher when only br-lan requires
the filtering. This setting would not be possible with
644-bridge_optimize_netfilter_hooks.patch applied and thus can only be
compared with brfilter-global and filtering enabled for all interfaces.
Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 46835
