When running as KVM or Xen guest, packets may be received with
incomplete checksum[1]. Patch adopted from Fedora dhcp package.
[1]. http://article.gmane.org/gmane.linux.kernel/1003853
Signed-off-by: Alexey I. Froloff <raorn@raorn.name>
SVN-Revision: 33475
Use a patch from gentoo which allows building libmount even
if the C library does not supports %ms/%as.
See: https://bugs.gentoo.org/406303
SVN-Revision: 33270
It is required by sysupgrade on some boards, and it is only
working if a suitable /etc/fw_env.config is present on the
target board. If an user creates such configuration he/she
should know that it is a dangerous tool. The devices can
be bricked in several other ways anyway.
SVN-Revision: 33265
This is fix an issue with dnsmasq's RA that does not set the "on-link" bit, making all local IPv6 traffic go to the router then to the destination host, not directly to each other.
patch is from dnsmasq git
SVN-Revision: 33216
It seems that the comgt package does not handle the Huawei 3G USB dongle E176 correctly (and probably other Huawei dongles too). My dongle appears as ID 12d1:1001 Huawei Technologies Co., Ltd. E620 USB Modem and 3G/UMTS
connections work well. However, no connection is established if only 2G/GPRS is available: the pppd chat script fails with NO CARRIER although the dongle is registered to the network (via 2G). As outlined in this wiki or this
blog, Huawei chips use the AT^SYSCFG command to set 2G or 3G mode, which is not implemented in comgt at the moment. Thus I wrote a patch for /lib/network/3g.sh which adds support for the "service" option in the network
configuration with Huawei dongles. By default (if no "service" option is specified) also 2G is used when 3G is unavailable. The Huawei dongle is detected analogously to other chips (the output of gcom -d /dev/ttyUSB0 -s
/etc/gcom/getcardinfo.gcom is scanned for huawei).
Some further information: The AT^SYSCFG command seems to be respected only once after the dongle is attached (or after the host is powered up). Resetting the dongle seems to render the serial port unusable in some cases.
However, the patch sets a useful mode by default which should cover most use cases (3G preferred, but 2G allowed) and if 3G-only or 2G-only mode is required the device can be power cycled.
SVN-Revision: 33212
This takes the device_id and subsystem_id from the EEPROM, I'll add
the info for other Rt3xxx chips in the next days.
[jow: minor whitespace changes]
Signed-off-by: Daniel Golle <dgolle@allnet.de>
SVN-Revision: 33199
These are two Atheros modules we are using, just for the cosmetics to make them
show up with the proper names in LuCI.
Signed-off-by: Daniel Golle <dgolle@allnet.de>
SVN-Revision: 33186
rssileds is a small user-space process to control LEDs by polling the
signal quality reported by a WiFi interface. By using the iwinfo library,
rssileds is independent of the WiFi driver used.
It supports pwm controlled LEDs and may by used to nicely fade through
all colors in real-time of the rainbow while only wasting very little CPU
time and a small constant amount of system memory.
An example configuration for the ALL0258N will follow in the next patch.
This is a slightly improved version of rssileds, now quality values are
in percent and stuff is written to syslog.
Signed-off-by: Daniel Golle <dgolle@allnet.de>
SVN-Revision: 33163
When the underlying /etc/config/wireless got changed since the last "wifi up"
has been performed, the uci vap ifname state vars become inconsistent on a
subsequent "wifi up" and multiple vaps get mapped to the same ifnames which
confuses the gui and other processes relying on them.
For now call an explicit "wifi down" prior to each "wifi up" which will clear
up the state accordingly.
SVN-Revision: 33076
Give nas interface time to setup them selves before configuring them.
Should fix#11826.
Contributed by T-Labs, Deutsche Telekom Innovation Laboratories
Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
SVN-Revision: 33074
The structs defined in these header files should be the same as the
structes used by the platform code otherwise it will cause strange
runtime errors. By removing them from compat-wireless we will get a
compile error instead of a runtime error if they are out of sync.
SVN-Revision: 33006
* remove use_et, all supported drivers are using mii ioctls
* remove robo.phy_addr, phy_addr was always set to ROBO_PHY_ADDR.
* remove support for old kernel versions
* do_ioctl is allways called with buff == NULL
* use if_mii
SVN-Revision: 33002
When standalone ldd is selected in config the binary is installed in /usr/bin/ldd.
Signed-off-by: Michael Heimpold <mhei@heimpold.de>
SVN-Revision: 32891
I did not add the last patch, because it add some stubs for the flash drivers and our patches have to be changed for that, if the flash support in bcma is able to do something I will integrate it.
This also contains 3 additional patches fixing some bugs in bcma.
SVN-Revision: 32880
With the BRCM47xx image I have built (Linksys E3000), there are additional
writes following the completion of "mtd_replace_jffs2" which invalidate the
'trx_fixup' performed by 'mtd_replace_jffs2'. Moving the 'trx_fixup' to somewhere
after all writes have completed fixes the problem. I also noticed that 'erasesize'
used to compute 'block_offset' in 'mtd_fixtrx' is used before it is computed by
'mtd_check_open'; moving the call to 'mtd_check_open' up a few lines fixes this.
Unlike 'mtd_fixtrx', 'trx_fixup' appears to assume that the TRX header is always at
offset 0; which may be the cause of the problem described in Ticket #8960.
Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 32866
This patch adds the ext4 filesystem support to mountd. Tested with ext2, ext3
and ext4 partitions.
Signed-off-by: Jaroslaw Swierczynski <jarek1701@gmail.com>
SVN-Revision: 32818
Forgot to do make package/mac80211/update, so the previously posted patch wasn't
in sync with compat-wireless. Now fixed.
Signed-off-by: Daniel Golle <dgolle@allnet.de>
SVN-Revision: 32817
iConnect board tested by:
Tim Fletcher <tim@night-shade.org.uk>
Wojciech Dubowik <wojciech.dubowik@neratec.com>
DockStar board tested by:
Martin Mueller <mm@sig21.net>
RaidSonic ICY BOX NAS6210 board tested by:
Luka Perkov <uboot@lukaperkov.net>
SheevaPlug was not tested but support for SheevaPlug is taken from upstream
uboot and it is not reported to be broken there.
SVN-Revision: 32717
Here is a newer version of the patch:
Enable the real ALSA midi interface (seq) in the kernel. The existing
rawmidi interface is still present. This allows for the alsa midi
utilities to work.
Signed-off-by: Adam Goode <adam@spicenitz.org>
[b_tsiligiannis: replaced $(LINUX_KMOD_SUFFIX) with .ko]
Signed-off-by: Vasilis Tsiligiannis <b_tsiligiannis@silverton.gr>
SVN-Revision: 32712
- remove unused variables
- simply ignore command line args which belong to not enabled features
- resolve peer address at accept() time, should solve (#11850)
- remove floating point operations where possible
SVN-Revision: 32704
- avoid closing descriptors before removing them from uloop (#11755, #11830)
- do not auto-initialize ubus if no prefix is set (#11832)
- remove extraneous client context pointer from cgi and lua states
- code cleanups and debug message changes
SVN-Revision: 32651
Add a config list to the 'config dnsmasq' section to specify fixed DNS
addresses.
For example:
config dnsmasq:
[snip]
list address '/example.com/192.168.0.1'
will result in the argument '-A /example.com/192.168.0.1' to the dnsmasq
options. This configures dnsmasq to return the specified IP for any
queries to '*.example.com' names.
Useful for overriding lookups to a range of DNS names.
[jow: "append args" -> "xappend", "-A" -> "--address"]
Signed-off-by: Jeremy Kerr <jk@ozlabs.org>
SVN-Revision: 32624
-T, --local-ttl=<time>
When replying with information from /etc/hosts or the DHCP leases
file dnsmasq by default sets the time-to-live field to zero, meaning
that the requestor should not itself cache the information. This is
the correct thing to do in almost all situations. This option allows a
time-to-live (in seconds) to be given for these replies. This will
reduce the load on the server at the expense of clients using stale
data under some circumstances.
[jow: change -T to --local--ttl to conform with the other options]
Signed-off-by: Andrew Byrne <openwrt@andy.id.au>
openwrt-devel mailing list
openwrt-devel@lists.openwrt.orghttps://lists.openwrt.org/mailman/listinfo/openwrt-devel
SVN-Revision: 32623
Set PATH in non-interactive logins to include /sbin paths,
so to be consistent with what is currently set in /etc/profile
for interactive shells.
[jow: reapply with current patch level, fix inner patch, refresh]
Signed-off-by: Gui Iribarren <gui@altermundi.net>
SVN-Revision: 32620
When mtd alters the fis partition table it assumes that the first partition
table entry also is the first logical parition table entry. For instance our
table could look like this (irrelevant partitions put aside):
* vmlinux.bin.l7 0xA8710000
* rootfs 0xA8030000
Here mtd would assume vmlinux.bin.l7 being the first partition and use its
address to calculate the size and offset which ultimately leads to a broken
partition table.
This patch alters the behavior by checking what partition has the smaller
address to do the calculations based on that address.
Signed-off-by: Marek Lindner <lindner_marek@yahoo.de>
SVN-Revision: 32601
The offset parameter can be used to write the data at the offset
instead of writing it to the beginning of the partition.
Signed-off-by: Marek Linder <lindner_marek@yahoo.de>
SVN-Revision: 32600
This adds a new boolean option, fqdn, to the "config dnsmasq" section of
/etc/config/dhcp. The default is off. When set on, it enables the dhcp-fqdn
option to dnsmasq. dhcp-fqdn causes dnsmasq's DNS server to not resolve
unqualifed local hostnames. The "domain" option is required when using "fqdn".
Local hostnames will remain available for lookup using fully-qualified names.
Signed-off-by: Mark Mentovai <mark@moxienet.com>
SVN-Revision: 32569
dnsmasq currently permits dhcp_options to be specified only in "config dhcp"
sections of /etc/config/dhcp. When dnsmasq is providing DHCP service for
multiple subnets and there are multiple "config dhcp" sections without "option
ignore", it makes sense to allow dhcp_options that should apply globally in
the "config dnsmasq" section of /etc/config/dhcp. dhcp_option is a list option.
[jow: rework patch to apply after dhcp-option-force handling got introduced]
Signed-off-by: Mark Mentovai <mark@moxienet.com>
SVN-Revision: 32568
Extroot works fine when the target device is specified by a path. It fails
however if the device is specified by UUID (the target partition gets mounted
much later by hotplug hooks). This is because the blkid command is no longer
compiled into BusyBox (since changeset [1]) so it's unavailable for the
preinit phase.
The closest bug report I was able to find is [2], although the reporting person
mentions that /tmp/overlay-disabled showed up which wasn't there in my case.
This patch sets PATH and LD_LIBRARY_PATH environment variables so that the
blkid command installed on the target device can be used by that particular
preinit script.
[1] https://dev.openwrt.org/changeset/26245
[2] https://dev.openwrt.org/ticket/10653
Signed-off-by: Jaroslaw Swierczynski <jarek1701@gmail.com>
SVN-Revision: 32567
l2tp_ppp needs to be loaded after pppox, otherwise it ends up like this:
l2tp_ppp: Unknown symbol pppox_ioctl (err 0)
...
during boot.
I also fixed the dependency, it should be pppox rather than pppoe.
Signed-off-by: Daniel Golle <dgolle@allnet.de>
SVN-Revision: 32562
A year of testing in the cerowrt project shows not using timestamps
to be a very bad idea in nearly any TCP at speeds above a few Mbit.
Lastly sack/dsack help on recovery from larger amounts of packet
loss.
SVN-Revision: 32513
Hi,
yes, it is true...
In the standard, unpatched trunk is zlib_inflate.ko compiled, but not included
in any package... So, my previous version was functional, but with system bug.
Here is fixed patch.
On Wed, Jun 13, 2012 at 05:00:02PM +0200, Jo-Philipp Wich wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> - -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Hi.
>
> Shouldn't you also add CONFIG_ZLIB_DEFLATE to KCONFIG then?
>
> ~ Jow
> - -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.12 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
>
> iEYEARECAAYFAk/YqrcACgkQdputYINPTPM6dgCfYDgzK1XHiKDJNSdc/+HgIoRp
> HSgAoKdUxcqXzHqTLiyEkiQqCnDuuVmu
> =0DUX
> - -----END PGP SIGNATURE-----
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.12 (GNU/Linux)
> Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/
>
> iEYEARECAAYFAk/YqvIACgkQdputYINPTPNMagCePNwPSYHfoCd5eXywQ+sTATqQ
> 2CQAoJW/Fez+DqflHlJVcvng/LvsfrCm
> =s6B0
> -----END PGP SIGNATURE-----
> _______________________________________________
> openwrt-devel mailing list
> openwrt-devel@lists.openwrt.org
> https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Signed-off-by: Michal Heppler <mhepp@ics.muni.cz>
SVN-Revision: 32444
Hi,
I found that openssl did not compile on the uml target under x86_64. The
attached patch should
correct this and is working for me. Is this the right way to do it?
thanks,
Thomas
SVN-Revision: 32443
kmod-ipt-nathelper-extra is missing the package nf_conntrack_broadcast.ko
if it is not included into the kmod-ipt-nathelper-extra packge the modules
nf_conntrack_snmp and nf_nat_snmp_basic cant get loaded:
[ 44.500000] nf_conntrack_snmp: Unknown symbol nf_conntrack_broadcast_help (err 0)
[ 44.664000] nf_nat_snmp_basic: Unknown symbol nf_nat_snmp_hook (err 0)
Signed-off-by: Peter Wagner <tripolar@gmx.at>
SVN-Revision: 32434
This patch adds support for manually configuring 6rd tunnels. It depends on
the netifd patches I sent earlier, which add 6rd support.
A basic interface configuration looks like:
config interface 'wan6'
option proto '6rd'
option peeraddr '192.0.2.1'
option ip6prefix '2123::'
option ip6prefixlen '16'
option ip4prefixlen '0'
Where ip4prefixlen is optional and actually defaults to 0, which would use all
bits of the IPv4 in the calculated IPv6 subnet.
I believe it should be possible to configure a regular 6to4 tunnel using this,
and that we may want to merge the two eventually, but there are some larger
differences between the two at the moment:
- 6rd addresses can be more difficult to calculate. My ISP, for example, has
a setup with a v6 mask of 43 bits, and a v4 mask of 19.
- 6to4 has support for configuring radvd. This is something we want, of
course, but it seems best to deal with this in a separate patch.
Just creating a new package looked like the quickest way to get this in.
This work is based on the 6in4 package, and work by Stijn Tintel.
Signed-off-by: Stéphan Kochen <stephan@kochen.nl>
SVN-Revision: 32431
Florian Fainelli
Hauke Mehrtens
Felix Fietkau
Jo-Philipp Wich
Gabor Juhos
John Crispin
Jonas Gorski
Vasilis Tsiligiannis
Hamish Guthrie
Luka Perkov
Travis Kemen