In my machine, the file is created with arptable_filter before
arp_tables. This can't happen because the former depends on the
latter.
So, instead of rely on wildcards (*.ko), set by hand the files to
be loaded in the correct order.
Signed-off-by: Jonh Wendell <jonh.wendell@oiwifi.com.br>
SVN-Revision: 36512
libiptc.pc depends on libip[4|6]tc.pc, thus all of those need to be
installed.
Should fix collectd build and thus #13146; which should make collectd
appear in snapshots again.
Signed-off-by: Danny Baumann <dannybaumann@web.de>
SVN-Revision: 36509
I recently picked up a WNDR3700 to put OpenWRT on, and only after tearing into the box did I find it
was one of the v3 boards, with poor OpenWRT support. This patch should add the board detection and
LED/button control to the broadcom-diag module, and should generate a netgear .chk image that the
bootloader and stock firmware will accept.
The changes to the broadcom-diag module are more than a few lines because the WNDR3700v3 is driving
its LEDs through an HC164 8-bit shift register.
Signed-off-by: Owen Kirby <osk@exegin.com>
SVN-Revision: 36482
Hardcoding channel number to 11 is a bad idea, it may be even not
available on some devices. If there are two single-band radios, the
one for 5GHz doesn't have 11 channel.
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 36479
It's needed for dealing with certain UART-to-USB adapters built into
e.g. weather stations, and increases the hid module package only
slightly (e.g. on ar71xx: 35k -> 37k).
Signed-off-by: Danny Baumann <dannybaumann@web.de>
SVN-Revision: 36415
Package up the Microchip MCP23S08 I/O expander driver. Needed on the
CalAmp LMU5000 board.
Signed-off-by: Adam Porter <porter.adam@gmail.com>
[jogo@openwrt.org: fix subject]
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 36393
- fixes memory leak in libuci when trying to open not existing configurations
- fixes memory leak in libuci-lua when querying not existing sections or options
- fixes handling of filenames with leading underscore
SVN-Revision: 36377
If the first byte is available in the address mask, use only that one -
set the local bit and xor it with the id << 2. This ensures that there
are no hardware BSSID & BSSID-mask conflicts with devices that have
almost the same MAC address with just a small offset. The MAC address
conflict has been observed in a deployment with some devices from the
same batch when running with multiple interfaces.
If only some bits of the last byte are available, xor the id onto the
last MAC address byte (relevant mostly for Ralink devices).
In other cases (should not happen at this point), use the previous MAC
address offset calculation but without the local bit.
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 36353
