openwrt/svn-archive/archive.git
5 years agozram-swap: fix dependencies
Hauke Mehrtens [Wed, 13 Nov 2013 18:42:45 +0000 (18:42 +0000)]
zram-swap: fix dependencies

swapon/swapoff are no longer provided by swap-utils; they are provided
by block-mount instead.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38795

5 years agobrcm47xx: swap also early boot console if necessary.
Hauke Mehrtens [Wed, 13 Nov 2013 18:38:18 +0000 (18:38 +0000)]
brcm47xx: swap also early boot console if necessary.

Some devices are using tty1 as the default console, also swap the early
boot console in such case. This is needed for the WGT634U.

Thank you Russell for reporting this.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 38794

5 years agokernel: b53: fix untagged shift for BCM5365
Hauke Mehrtens [Wed, 13 Nov 2013 18:04:39 +0000 (18:04 +0000)]
kernel: b53: fix untagged shift for BCM5365

The BCM5365 needs a shift of 7 bits and not 6 bits like the BCM5325 for
the untagged ports.

Thank you Russell for reporting this and testing the patch.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 38793

5 years ago[mac80211]: upgrade linux-firmware git revision, bump package version
Imre Kaloz [Wed, 13 Nov 2013 14:47:31 +0000 (14:47 +0000)]
[mac80211]: upgrade linux-firmware git revision, bump package version

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 38792

5 years ago[mac80211]: add another symbol for TI chipsets
Imre Kaloz [Wed, 13 Nov 2013 14:46:21 +0000 (14:46 +0000)]
[mac80211]: add another symbol for TI chipsets

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 38791

5 years agoprocd: remove obselete script install reference
John Crispin [Wed, 13 Nov 2013 12:15:06 +0000 (12:15 +0000)]
procd: remove obselete script install reference
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38790

5 years agoprocd: fixup 2 wrong option types
John Crispin [Wed, 13 Nov 2013 11:07:24 +0000 (11:07 +0000)]
procd: fixup 2 wrong option types

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38789

5 years agoprocd: options with default values are not parsed properly in the ubus register code
John Crispin [Wed, 13 Nov 2013 10:57:16 +0000 (10:57 +0000)]
procd: options with default values are not parsed properly in the ubus register code

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38788

5 years agoprocd: convert services to the new validation api
John Crispin [Wed, 13 Nov 2013 10:49:41 +0000 (10:49 +0000)]
procd: convert services to the new validation api

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38787

5 years agoprocd: update to latest git head
John Crispin [Wed, 13 Nov 2013 10:49:25 +0000 (10:49 +0000)]
procd: update to latest git head

this includes the first wip version of the uci validation backend

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38786

5 years ago[package]: uboot-sunxi: add U-boot builds for a couple sunxi boards
Zoltan Herpai [Wed, 13 Nov 2013 10:49:22 +0000 (10:49 +0000)]
[package]: uboot-sunxi: add U-boot builds for a couple sunxi boards

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
SVN-Revision: 38785

5 years agoramips: select kmod-rt2800-soc by default (if available)
Felix Fietkau [Wed, 13 Nov 2013 10:29:10 +0000 (10:29 +0000)]
ramips: select kmod-rt2800-soc by default (if available)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 38784

5 years agomac80211: update to wireless-testing 2013-11-05
Felix Fietkau [Tue, 12 Nov 2013 22:11:33 +0000 (22:11 +0000)]
mac80211: update to wireless-testing 2013-11-05

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 38783

5 years agohostapd: remove #ifdef MULTICALL around ap+sta support code (to support separately...
Felix Fietkau [Tue, 12 Nov 2013 22:11:14 +0000 (22:11 +0000)]
hostapd: remove #ifdef MULTICALL around ap+sta support code (to support separately installed supplicant+hostapd)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 38782

5 years agomxs: - update uboot-mxs to 3.10
Zoltan Herpai [Tue, 12 Nov 2013 14:39:32 +0000 (14:39 +0000)]
mxs: - update uboot-mxs to 3.10
 - enable building uboot for imx23 along sbImage by default
 - (doc updated)

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
SVN-Revision: 38781

5 years agomxs: bump kernel to 3.12
Zoltan Herpai [Tue, 12 Nov 2013 12:48:10 +0000 (12:48 +0000)]
mxs: bump kernel to 3.12
 - remove DCP patch as it's in mainline now (dts patch will stay)
 - refresh Duckbill patch
 - update kmod-usb-chipidea

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
SVN-Revision: 38780

5 years agoAdd patch to fix ipset timeout setting problem
Jo-Philipp Wich [Tue, 12 Nov 2013 11:01:11 +0000 (11:01 +0000)]
Add patch to fix ipset timeout setting problem

Fixes #14325

Signed-off-by: Nikolay Martynov <mar.kolya@gmail.com>
SVN-Revision: 38779

5 years agoUpdate ipset to version 6.20.1
Jo-Philipp Wich [Tue, 12 Nov 2013 11:00:49 +0000 (11:00 +0000)]
Update ipset to version 6.20.1

Signed-off-by: Nikolay Martynov <mar.kolya@gmail.com>
SVN-Revision: 38778

5 years agomxs: add basic board runtime detection support
Zoltan Herpai [Tue, 12 Nov 2013 10:19:26 +0000 (10:19 +0000)]
mxs: add basic board runtime detection support

Signed-off-by: Michael Heimpold <mhei@heimpold.de>
Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
SVN-Revision: 38777

5 years agouboot-envtools: add support for mxs target
Zoltan Herpai [Tue, 12 Nov 2013 10:07:03 +0000 (10:07 +0000)]
uboot-envtools: add support for mxs target

Signed-off-by: Michael Heimpold <mhei@heimpold.de>
SVN-Revision: 38776

5 years agomxs: duckbill: adjust leds to final hardware
Zoltan Herpai [Tue, 12 Nov 2013 09:38:14 +0000 (09:38 +0000)]
mxs: duckbill: adjust leds to final hardware

Signed-off-by: Michael Heimpold <mhei@heimpold.de>
SVN-Revision: 38775

5 years agomxs: rename uboot-imx23 to uboot-mxs
Zoltan Herpai [Tue, 12 Nov 2013 01:34:30 +0000 (01:34 +0000)]
mxs: rename uboot-imx23 to uboot-mxs

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
SVN-Revision: 38774

5 years agomxs: fix loadaddr for uImages
Zoltan Herpai [Tue, 12 Nov 2013 01:25:53 +0000 (01:25 +0000)]
mxs: fix loadaddr for uImages

Signed-off-by: Harald Geyer <harald@ccbib.org>
SVN-Revision: 38773

5 years agokernel: add new core sound module snd-pcm-dmaengine.ko for 3.12 kernels and above
Zoltan Herpai [Tue, 12 Nov 2013 01:09:15 +0000 (01:09 +0000)]
kernel: add new core sound module snd-pcm-dmaengine.ko for 3.12 kernels and above

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
SVN-Revision: 38772

5 years ago[package] base-files: don't enter failsafe if triggered by switch
Luka Perkov [Mon, 11 Nov 2013 23:34:11 +0000 (23:34 +0000)]
[package] base-files: don't enter failsafe if triggered by switch

Fixes failsafe issues with switches after r38725.

Signed-off-by: Luka Perkov <luka@openwrt.org>
SVN-Revision: 38770

5 years agobrcm47xx: fix netconfig for inverted vlan?ports in nvram
Hauke Mehrtens [Mon, 11 Nov 2013 23:04:47 +0000 (23:04 +0000)]
brcm47xx: fix netconfig for inverted vlan?ports in nvram

This was observed on the Belkin Share F7D7302.

SVN-Revision: 38769

5 years agobroadcom-diag: add Belkin F7D3302/F7D7302
Hauke Mehrtens [Mon, 11 Nov 2013 23:04:23 +0000 (23:04 +0000)]
broadcom-diag: add Belkin F7D3302/F7D7302

Signed-off-by: Cody P Schafer <devel@codyps.com>
SVN-Revision: 38768

5 years agolinux/brcm47xx : Belkin F7Dx30x board detection
Hauke Mehrtens [Mon, 11 Nov 2013 23:04:00 +0000 (23:04 +0000)]
linux/brcm47xx : Belkin F7Dx30x board detection

boardnum is unstable, so allow using NULL to make it unchecked.

A method for differentiating F7D3301 from F7D3302 using nvram is
unknown at this point. Reading the first 32bits of magic from the
"linux" mtd partition would do the trick, but these 2 are similar
enough that we don't need separate led/button definitions. Not knowing
which one you have is only a usability issue for upgrading the device:
they expect different magic numbers for their trx images.

Signed-off-by: Cody P Schafer <devel@codyps.com>
SVN-Revision: 38767

5 years agobase-files: don't kill nas for sysupgrade
Hauke Mehrtens [Mon, 11 Nov 2013 22:12:58 +0000 (22:12 +0000)]
base-files: don't kill nas for sysupgrade

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38766

5 years agobroadcom-wl: version roll
Hauke Mehrtens [Mon, 11 Nov 2013 22:05:39 +0000 (22:05 +0000)]
broadcom-wl: version roll

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38765

5 years agobroadcom-wl: support 'wds' interfaces on 'wl1' device
Hauke Mehrtens [Mon, 11 Nov 2013 22:05:20 +0000 (22:05 +0000)]
broadcom-wl: support 'wds' interfaces on 'wl1' device

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38764

5 years agobroadcom-wl: support probe of 'wds' devices
Hauke Mehrtens [Mon, 11 Nov 2013 22:04:53 +0000 (22:04 +0000)]
broadcom-wl: support probe of 'wds' devices

'nas' requires probe of 'wds' devices to be successful;
so don't fail them.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38763

5 years agobroadcom-wl: fix crash when starting multiple virtual interfaces
Hauke Mehrtens [Mon, 11 Nov 2013 22:04:26 +0000 (22:04 +0000)]
broadcom-wl: fix crash when starting multiple virtual interfaces

When enabling multiple VIFS, the driver sometimes crashes.  The frequency
of the crash increases as more VIFS are enabled.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38762

5 years agobroadcom-wl: perform additional cleanup on wl module unload
Hauke Mehrtens [Mon, 11 Nov 2013 22:03:58 +0000 (22:03 +0000)]
broadcom-wl: perform additional cleanup on wl module unload

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38761

5 years agokernel: of_i2c is moved into i2c-core in 3.12, add kernel version check into modules...
Zoltan Herpai [Mon, 11 Nov 2013 22:03:35 +0000 (22:03 +0000)]
kernel: of_i2c is moved into i2c-core in 3.12, add kernel version check into modules/i2c.mk

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
SVN-Revision: 38760

5 years agobroadcom-wl: support more virtual interfaces
Hauke Mehrtens [Mon, 11 Nov 2013 22:03:25 +0000 (22:03 +0000)]
broadcom-wl: support more virtual interfaces

Read the number of virtual interfaces to support from the device
capabilities; as some newer devices can support up to 16.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38759

5 years agobroadcom-wl: improve device cleanup
Hauke Mehrtens [Mon, 11 Nov 2013 22:02:45 +0000 (22:02 +0000)]
broadcom-wl: improve device cleanup

Move all device cleanup to "disable_broadcom".

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38758

5 years agobroadcom-wl: add support for setting H/W address
Hauke Mehrtens [Mon, 11 Nov 2013 22:02:14 +0000 (22:02 +0000)]
broadcom-wl: add support for setting H/W address

Some devices initialize with a default address (common to all H/W).
The address needs to be changeable so that a unique address can be
assigned to each AP.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38757

5 years agobroadcom-wl: eliminate 'invalid command' when setting ssid
Hauke Mehrtens [Mon, 11 Nov 2013 22:01:32 +0000 (22:01 +0000)]
broadcom-wl: eliminate 'invalid command' when setting ssid

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38756

5 years agobroadcom-wl: clean up variable definitions (broadcom.sh)
Hauke Mehrtens [Mon, 11 Nov 2013 22:01:12 +0000 (22:01 +0000)]
broadcom-wl: clean up variable definitions (broadcom.sh)

Specify variables as 'local' where appropriate

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38755

5 years agobroadcom-wl: simplify processing of 'hwmode'
Hauke Mehrtens [Mon, 11 Nov 2013 22:00:39 +0000 (22:00 +0000)]
broadcom-wl: simplify processing of 'hwmode'

Default gmode to "GOnly" unless explicitly overwridden.  This corrects
processing for 11ng to specify "GOnly" for gmode instead of "Auto"
(which translated to "bg").

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38754

5 years agobroadcom-wl: shutdown 'nas' when 'wl' device is disabled
Hauke Mehrtens [Mon, 11 Nov 2013 21:59:58 +0000 (21:59 +0000)]
broadcom-wl: shutdown 'nas' when 'wl' device is disabled

'nas' must be shutdown independently for each 'wl' device to properly
support devices with multiple radios.  Do the shutdown when the device
is disabled instead of when it is brought up.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38753

5 years agobroadcom-wl: use a single instance of 'nas' per 'wl' device
Hauke Mehrtens [Mon, 11 Nov 2013 21:59:34 +0000 (21:59 +0000)]
broadcom-wl: use a single instance of 'nas' per 'wl' device

A single instance of 'nas' can support multiple interfaces; there's
no need to run multiple instances.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38752

5 years agobroadcom-wl: fix whitespace handling
Hauke Mehrtens [Mon, 11 Nov 2013 21:59:07 +0000 (21:59 +0000)]
broadcom-wl: fix whitespace handling

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38751

5 years agobroadcom-wl: update description for recently added
Hauke Mehrtens [Mon, 11 Nov 2013 21:58:42 +0000 (21:58 +0000)]
broadcom-wl: update description for recently added

Update descriptions for 'phytype' and 'band'.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38750

5 years agobroadcom-wl: fix band reporting
Hauke Mehrtens [Mon, 11 Nov 2013 21:58:19 +0000 (21:58 +0000)]
broadcom-wl: fix band reporting

Modify the band reporting for broadcom devices to correctly report
HW capability.  Currently, 802.11na devices report "bgn"; but should be
reporting "an".

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38749

5 years agobroadcom-wl: change interface names
Hauke Mehrtens [Mon, 11 Nov 2013 21:57:26 +0000 (21:57 +0000)]
broadcom-wl: change interface names

'netifd' treats all devices with a "." in their name as vlan devices.
Modify the name of virtual/WDS interfaces created by broadcom-wl to use
a "-" instead of a ".", so they will be treated as simple devices
by 'netifd'.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38748

5 years agobroadcom-wl: fix compiler warning
Hauke Mehrtens [Mon, 11 Nov 2013 21:56:45 +0000 (21:56 +0000)]
broadcom-wl: fix compiler warning

Delete the return value (nvram_exit is declared with a 'void' return).

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38747

5 years agoiwinfo: don't report associations for broadcom wds interfaces
Hauke Mehrtens [Mon, 11 Nov 2013 21:54:55 +0000 (21:54 +0000)]
iwinfo: don't report associations for broadcom wds interfaces

The associations list for Broadcom WDS interfaces are duplicate to those
reported by the corresponding AP interface; so there is no need to report
them again.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38746

5 years agoiwinfo: fix hwmodelist reporting for broadcom wl
Hauke Mehrtens [Mon, 11 Nov 2013 21:54:25 +0000 (21:54 +0000)]
iwinfo: fix hwmodelist reporting for broadcom wl

Modify the hwmodelist reporting for broadcom devices to include
proper reporting of 802.11n support.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 38745

5 years agobrcm47xx: b44: fix logic in BUG_ON()
Hauke Mehrtens [Mon, 11 Nov 2013 21:23:29 +0000 (21:23 +0000)]
brcm47xx: b44: fix logic in BUG_ON()

It is a problem when phydev is not set at this position and not the other way around.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 38744

5 years agouhttpd: return after processing -d switch, only set ubus prefix if not already defined
Jo-Philipp Wich [Mon, 11 Nov 2013 19:14:01 +0000 (19:14 +0000)]
uhttpd: return after processing -d switch, only set ubus prefix if not already defined

SVN-Revision: 38743

5 years agobase-files: fix typo in rfkill script
John Crispin [Mon, 11 Nov 2013 18:12:30 +0000 (18:12 +0000)]
base-files: fix typo in rfkill script

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38742

5 years agouhttp: add ubus.default
John Crispin [Mon, 11 Nov 2013 18:12:26 +0000 (18:12 +0000)]
uhttp: add ubus.default

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38741

5 years agoixp4xx: fix usb on cambria boards by using the ehci-platform driver
Felix Fietkau [Mon, 11 Nov 2013 16:51:01 +0000 (16:51 +0000)]
ixp4xx: fix usb on cambria boards by using the ehci-platform driver

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 38740

5 years agopppd: don't call modprobe at all from pppoatm
John Crispin [Mon, 11 Nov 2013 13:40:28 +0000 (13:40 +0000)]
pppd: don't call modprobe at all from pppoatm

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38737

5 years agorevert 94e44643a8ce1d7991376b6ced586784b3f92500
John Crispin [Mon, 11 Nov 2013 13:40:24 +0000 (13:40 +0000)]
revert 94e44643a8ce1d7991376b6ced586784b3f92500

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38736

5 years agoramips: use gpio-button-hotplug in favour of the input variant
John Crispin [Mon, 11 Nov 2013 13:40:20 +0000 (13:40 +0000)]
ramips: use gpio-button-hotplug in favour of the input variant

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38735

5 years agoramips: make the rfkill switch work on fonera20n
John Crispin [Mon, 11 Nov 2013 13:40:17 +0000 (13:40 +0000)]
ramips: make the rfkill switch work on fonera20n

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38734

5 years agolantiq: DGN3500 fixups
John Crispin [Mon, 11 Nov 2013 13:15:08 +0000 (13:15 +0000)]
lantiq: DGN3500 fixups

Just fixing up a few things for the Netgear DGN3500.

Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
SVN-Revision: 38733

5 years agoFix modprobe path for pppoatm
John Crispin [Mon, 11 Nov 2013 13:15:04 +0000 (13:15 +0000)]
Fix modprobe path for pppoatm

Without netifd generates an error:

daemon.notice netifd: wan (1249): /bin/sh: /sbin/modprobe: not found

Signed-off-by: Cezary Jackiewicz <cezary.jackiewicz@gmail.com>
SVN-Revision: 38732

5 years agolantiq: dont install deu test manager
John Crispin [Mon, 11 Nov 2013 12:22:12 +0000 (12:22 +0000)]
lantiq: dont install deu test manager

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38731

5 years agolantiq: rename TDW8970 led adsl->dsl
John Crispin [Mon, 11 Nov 2013 12:19:26 +0000 (12:19 +0000)]
lantiq: rename TDW8970 led adsl->dsl

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38730

5 years agobase-files: update rfkill script to work with slider buttons
John Crispin [Mon, 11 Nov 2013 12:19:22 +0000 (12:19 +0000)]
base-files: update rfkill script to work with slider buttons

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38729

5 years agolantiq: update TDW8970.dts to support buttons properly
John Crispin [Mon, 11 Nov 2013 12:19:18 +0000 (12:19 +0000)]
lantiq: update TDW8970.dts to support buttons properly

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38728

5 years agolantiq: remove old hotplug handler
John Crispin [Mon, 11 Nov 2013 12:19:15 +0000 (12:19 +0000)]
lantiq: remove old hotplug handler

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38727

5 years agobase-files: add a generic wps button handler
John Crispin [Mon, 11 Nov 2013 12:19:10 +0000 (12:19 +0000)]
base-files: add a generic wps button handler

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38726

5 years agogpio-button-hotplug: add support for sliding switches
John Crispin [Mon, 11 Nov 2013 12:19:06 +0000 (12:19 +0000)]
gpio-button-hotplug: add support for sliding switches

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38725

5 years agouhttp: make the service auto respawn if it crashes
John Crispin [Mon, 11 Nov 2013 12:19:02 +0000 (12:19 +0000)]
uhttp: make the service auto respawn if it crashes

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38724

5 years agouhttp: make sure that the ubus module is loaded when when installed
John Crispin [Mon, 11 Nov 2013 12:18:58 +0000 (12:18 +0000)]
uhttp: make sure that the ubus module is loaded when when installed

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38723

5 years ago[omap]: add missing option, enable hw rng
Imre Kaloz [Mon, 11 Nov 2013 11:55:44 +0000 (11:55 +0000)]
[omap]: add missing option, enable hw rng

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 38722

5 years ago[crypto-hw-omap]: depend on the correct target
Imre Kaloz [Mon, 11 Nov 2013 11:51:54 +0000 (11:51 +0000)]
[crypto-hw-omap]: depend on the correct target

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 38721

5 years agoar71xx: Use KEY_RESTART key code for the reset button on WD My Net N600 boards
Gabor Juhos [Mon, 11 Nov 2013 08:10:02 +0000 (08:10 +0000)]
ar71xx: Use KEY_RESTART key code for the reset button on WD My Net N600 boards

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38720

5 years agoar71xx: add profile and build image for the TL-WA750RE TL-WA850RE range extender
Gabor Juhos [Mon, 11 Nov 2013 08:04:48 +0000 (08:04 +0000)]
ar71xx: add profile and build image for the TL-WA750RE TL-WA850RE range extender

Patch to add profile and build image for the tplink TL-WA750RE/WA850RE board

Signed-off-by: Martijn Zilverschoon <thefriedzombie@gmail.com>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38719

5 years agoar71xx: add user space support for the TL-WA750RE/WA850RE Range extender
Gabor Juhos [Mon, 11 Nov 2013 08:04:47 +0000 (08:04 +0000)]
ar71xx: add user space support for the TL-WA750RE/WA850RE Range extender

Patch to add user space support for the TL-WA750RE/WA850RE range extender

Signed-off-by: Martijn Zilverschoon <thefriedzombie@gmail.com>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38718

5 years agoar71xx: Add kernel support for the TP-Link WA750RE / WA850RE range extender
Gabor Juhos [Mon, 11 Nov 2013 08:04:46 +0000 (08:04 +0000)]
ar71xx: Add kernel support for the TP-Link WA750RE / WA850RE range extender

Patch to add kernel support for the TP-LINK WA750RE and the WA850RE
range extender

Signed-off-by: Martijn Zilverschoon <thefriedzombie@gmail.com>
[juhosg: rename and refresh kernel patch]
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38717

5 years agoar71xx: ag71xx: fix ag71xx_probe error path
Gabor Juhos [Mon, 11 Nov 2013 07:47:00 +0000 (07:47 +0000)]
ar71xx: ag71xx: fix ag71xx_probe error path

Signed-off-by: Catalin Patulea <cat@vv.carleton.ca>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38716

5 years agobcm53xx: make bgmac work
Hauke Mehrtens [Mon, 11 Nov 2013 00:15:00 +0000 (00:15 +0000)]
bcm53xx: make bgmac work

This makes bgmac work without sprom and refreshed the patches.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 38715

5 years agokernel: bgmac: make it send and receive some packages on BCM4708
Hauke Mehrtens [Mon, 11 Nov 2013 00:13:31 +0000 (00:13 +0000)]
kernel: bgmac: make it send and receive some packages on BCM4708

This adds some more code for bgmac core rev 4 and it now restarts all
cores when initializing the first one on BCM4708. I am just able to
send under 100 packages and then DMA TX does not work any more.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 38714

5 years agokernel: bcma: detect pci ID 0x4313
Hauke Mehrtens [Mon, 11 Nov 2013 00:08:09 +0000 (00:08 +0000)]
kernel: bcma: detect pci ID 0x4313

This pci ID is used by sprom less BCM4313 devices.
This is a backport from the mainline kernel.

This is part of #13551.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 38713

5 years agokernel: b53: add initial support for switches in BCM5301X
Hauke Mehrtens [Sun, 10 Nov 2013 23:29:59 +0000 (23:29 +0000)]
kernel: b53: add initial support for switches in BCM5301X

This is also known as BCM470{7,8,9}.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 38712

5 years agoprocd: reduce the number of unnecessary calls to /sbin/hotplug-call for tty devices
Felix Fietkau [Sun, 10 Nov 2013 15:06:28 +0000 (15:06 +0000)]
procd: reduce the number of unnecessary calls to /sbin/hotplug-call for tty devices

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 38705

5 years agonetifd: remove connect_time from /var/state, it is unused
Felix Fietkau [Sun, 10 Nov 2013 10:01:33 +0000 (10:01 +0000)]
netifd: remove connect_time from /var/state, it is unused

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 38704

5 years ago[cns3xxx]: add uImage image build
Imre Kaloz [Sun, 10 Nov 2013 09:57:07 +0000 (09:57 +0000)]
[cns3xxx]: add uImage image build

- add building of a standard uImage along with the combined kernel+rootfs
- remove 'old' image build - that was for a very old and obsolete bootloader

To update the bootloader for GW2387,GW2388,GW2391 (NOR Flash) from uboot:
put latest NOR bootloader on tftp server:

http://svn.gateworks.com/laguna/trunk/images/u-boot_nor.bin

 Laguna> setenv ipaddr <localip>
 Laguna> setenv ipaddr <serverip>
 Laguna> tftpboot 0x800000 laguna/u-boot-nor.bin
 Laguna> erase 0x10000000 +$(filesize)
 Laguna> cp.b 0x800000 0x10000000 $(filesize)
 Laguna> reset

To update the bootloader for GW2380,GW2382,GW2383 (SPI Flash) from uboot:
put latest SPI bootloader on tftp server:

http://svn.gateworks.com/laguna/trunk/images/u-boot_spi.bin

 Laguna> setenv ipaddr <localip>
 Laguna> setenv ipaddr <serverip>
 Laguna> tftpboot 0x800000 laguna/u-boot-spi.bin
 Laguna> erase 0x60000000 +$(filesize)
 Laguna> cp.b 0x800000 0x60000000 $(filesize)
 Laguna> reset

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
SVN-Revision: 38703

5 years ago[cns3xxx]: add pushbutton GPIO for GW2380/82/83
Imre Kaloz [Sun, 10 Nov 2013 09:55:17 +0000 (09:55 +0000)]
[cns3xxx]: add pushbutton GPIO for GW2380/82/83

Signed-off-by: Tim Harvey <tharvey@gateworks.com>
SVN-Revision: 38702

5 years agobroadcom-diag: fix sys button on Asus rt-n16
Hauke Mehrtens [Sat, 9 Nov 2013 20:09:21 +0000 (20:09 +0000)]
broadcom-diag: fix sys button on Asus rt-n16

This closes: #14427.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 38701

5 years agobrcm47xx: add detection for Linksys WRT54GS V1
Hauke Mehrtens [Sat, 9 Nov 2013 20:08:09 +0000 (20:08 +0000)]
brcm47xx: add detection for Linksys WRT54GS V1

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 38700

5 years agobrcm47xx: register ADM6996L switch
Hauke Mehrtens [Sat, 9 Nov 2013 19:29:46 +0000 (19:29 +0000)]
brcm47xx: register ADM6996L switch

This patch makes it possible to use adm6996.c on first generation
BCM47XX devices with ADM switches.

This was tested on a WRT54GS version 1.0, thank you Dirk Neukirchen for
the device.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 38699

5 years agokernel: adm6996: add support for ADM6996L and GPIO interface
Hauke Mehrtens [Sat, 9 Nov 2013 19:27:27 +0000 (19:27 +0000)]
kernel: adm6996: add support for ADM6996L and GPIO interface

This patch makes it possible to use adm6996.c on first generation
BCM47XX devices with ADM switches.
The GPIO bit banging protocol implementation was copied from the old
switch driver and adapted to this driver and changed to the mainline
kernel GPIO interface.
The ADM6996L is different from the ADM6996M which is supported, for
both specs are available in the Internet.

This was tested on a WRT54GS version 1.0, thank you Dirk Neukirchen for
the device.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 38698

5 years agoprocd: update to latest git head
John Crispin [Fri, 8 Nov 2013 21:53:37 +0000 (21:53 +0000)]
procd: update to latest git head

make sure that automatic respawning is turned off during upgrade

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 38695

5 years ago[omap]: switch to 3.12, enable support for AM33xx/OMAP3
Imre Kaloz [Fri, 8 Nov 2013 12:44:38 +0000 (12:44 +0000)]
[omap]: switch to 3.12, enable support for AM33xx/OMAP3

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 38692

5 years ago[omap]: rename the omap4 target to omap
Imre Kaloz [Fri, 8 Nov 2013 12:37:39 +0000 (12:37 +0000)]
[omap]: rename the omap4 target to omap

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 38691

5 years agobase-files: mtd_get_mac_ascii bugfix
Gabor Juhos [Fri, 8 Nov 2013 12:16:21 +0000 (12:16 +0000)]
base-files: mtd_get_mac_ascii bugfix

The mtd_get_mac_ascii utility function was broken. This fixes it.

  - Remove the superfluous include of /lib/functions.sh. The
    function is already in that file so it is pointless,
  - only use a variable if the whole key word matches,
  - don't try to process the MAC address if it is empty,
  - use 'tr' to canonicalize the MAC address,

Signed-off-by: Daniel Gimpelevich <daniel@gimpelevich.san-francisco.ca.us>
[juhosg: add more fixes, update commit message]
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38690

5 years agoar71xx: ag71xx: fix a race involving netdev registration
Gabor Juhos [Fri, 8 Nov 2013 08:17:54 +0000 (08:17 +0000)]
ar71xx: ag71xx: fix a race involving netdev registration

In particular, phy_connect before register_netdev. This is because
register_netdev runs the netdev notifiers, which can race with the rest of
the initialization in ag71xx_probe. In my case this manifested in two ways:

1) If ag71xx is compiled as a module and inserted after netifd has started,
   netifd is notified by register_netdev before the call to
   ag71xx_phy_connect. netifd tries to bring the interface up, which calls
   ag71xx_open, which in turn enters ag71xx_phy_start. This keys off
   ag->phy_dev (which is still NULL) and thinks this is a fixed-link board,
   and enters ag71xx_link_adjust. This looks at ag->speed which is not yet
   initialized and hits the BUG() in the switch (ag->speed) in
   ag71xx_link_adjust.

   This is the wrong code path for ag71xx_phy_start - my board has PHYs that
   need to be brought up with phy_start. Doing ag71xx_phy_connect before
   register_netdev ensures that ag->phy_dev is non-NULL before
   ag71xx_phy_start is ever called.

2) When ag71xx is built into the kernel, and netconsole is enabled, there
   is a gap in the initial burst of replayed printks right after the netdev
   comes up. My assumption is that netconsole is also triggered by a netdev
   notifier, and part of this printk burst happens before the call into
   ag71xx_phy_connect, so part of the burst is lost while the PHY comes up.
   This patch fixes the gap - all the printks before eth0 comes up are bursted
   in full when netconsole initializes.

ag71xx_phy_connect_xxx no longer runs with a registered netdev, so the
logging has been adjusted accordingly to avoid "unregistered net_device" or
"eth%d" messages in dmesg.

Signed-off-by: Catalin Patulea <cat@vv.carleton.ca>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38689

5 years agoar71xx: build image for the WD My Net Wi-Fi Range Extender
Gabor Juhos [Thu, 7 Nov 2013 22:31:54 +0000 (22:31 +0000)]
ar71xx: build image for the WD My Net Wi-Fi Range Extender

Patchwork: http://patchwork.openwrt.org/patch/4281/
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
[juhosg:
  - remove the Image/Build/CyberTANLZMA/loader macro, it is not used
  - move the MYNETREXT SingleProfile definition to the correct place]
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38688

5 years agoar71xx: add user-space support for the WD My Net Wi-Fi Range Extender
Gabor Juhos [Thu, 7 Nov 2013 22:31:53 +0000 (22:31 +0000)]
ar71xx: add user-space support for the WD My Net Wi-Fi Range Extender

Patchwork: http://patchwork.openwrt.org/patch/4281/
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38687

5 years agoar71xx: add kernel support for the My Net Wi-Fi Range Extender device
Gabor Juhos [Thu, 7 Nov 2013 22:31:52 +0000 (22:31 +0000)]
ar71xx: add kernel support for the My Net Wi-Fi Range Extender device

This patch adds a new device definition for a
Western Digital device. The hardware seems to
be based on Qualcomm Atheros DB120 design.

Patchwork: http://patchwork.openwrt.org/patch/4281/
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
[juhosg: refresh kernel patch]
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38686

5 years agotools: adapt addpattern for WD's Range Extender
Gabor Juhos [Thu, 7 Nov 2013 22:31:48 +0000 (22:31 +0000)]
tools: adapt addpattern for WD's Range Extender

A few things had to be changed to add support for these devices:
 - support code patterns with 8 characters
 - new board definition for the range extender device

Patchwork: http://patchwork.openwrt.org/patch/4280/
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38685

5 years agomac80211: mark rt2x00 library packages as hidden, there's no point in allowing users...
Felix Fietkau [Thu, 7 Nov 2013 20:59:22 +0000 (20:59 +0000)]
mac80211: mark rt2x00 library packages as hidden, there's no point in allowing users to enable them without the packages that depend on them

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 38684

5 years agoar71xx: image: add EOF marker to the WNDR4300 UBI image
Gabor Juhos [Thu, 7 Nov 2013 19:53:02 +0000 (19:53 +0000)]
ar71xx: image: add EOF marker to the WNDR4300 UBI image

Append EOF markers to the END of the UBI image
file. This forces the kernel to erase all blocks
after the marker even if those blocks are not
empty.

Additionally, the resulting image can be flashed
from the original web UI now, so use '-factory'
suffix for that.

Signed-off-by: Gabor Juhos <juhosg@openwrt.org>
SVN-Revision: 38683