openwrt/staging/pepe2k.git
9 years agokernel: bridge, multicast-to-unicast: fix echoes on STA
Felix Fietkau [Wed, 2 Sep 2015 09:23:59 +0000 (09:23 +0000)]
kernel: bridge, multicast-to-unicast: fix echoes on STA

Currently, multicast packets from an STA are sent to any according
multicast listener directly through the bridge multicast-to-unicast
feature. Unfortunately, so far this includes the originating STA, too,
resulting in multicast packets being echo'ed back to the originating STA
if it itself is a multicast listener for that group.

This behaviour breaks IPv6 duplicate address detection: An IPv6 Neighbor
Solicitation for IPv6 Duplicate Address Detection is being echo'ed back,
resulting in the host falsely detecting an address collision, which
makes the node unable to claim an IPv6 address and use IPv6 in general.

Mac80211 unfortunately only prevents the echoes for us for multicast
frames. For the multicast frames cast to a unicast destination we'll
need to take care of excluding the originator ourselves.

Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue>
SVN-Revision: 46765

9 years agomdns: fix having mulitple network entries in uci
Felix Fietkau [Wed, 2 Sep 2015 09:23:52 +0000 (09:23 +0000)]
mdns: fix having mulitple network entries in uci

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 46764

9 years agomap: add debug-code for mapcalc
Steven Barth [Tue, 1 Sep 2015 18:48:22 +0000 (18:48 +0000)]
map: add debug-code for mapcalc

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 46763

9 years agopolarssl: bump to 1.3.12
Steven Barth [Tue, 1 Sep 2015 18:48:15 +0000 (18:48 +0000)]
polarssl: bump to 1.3.12

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 46762

9 years agoodhcpd: fix incorrect address assignment for DHCPv6
Steven Barth [Tue, 1 Sep 2015 13:31:00 +0000 (13:31 +0000)]
odhcpd: fix incorrect address assignment for DHCPv6

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 46761

9 years agolantiq: remove dead EASY33016 image recipe
Jonas Gorski [Sun, 30 Aug 2015 12:18:22 +0000 (12:18 +0000)]
lantiq: remove dead EASY33016 image recipe

Support for lantiq_svip_be has been removed a while ago, so EASY33016
images weren't buildable anymore. Remove the recipes as well as gzip
compressed kernel support, as EASY33016 was the last user of it.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46760

9 years agoramips: disable the openwrt commandline hack
Jonas Gorski [Sun, 30 Aug 2015 12:18:14 +0000 (12:18 +0000)]
ramips: disable the openwrt commandline hack

We don't make use of it, so no need to have it enabled.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46759

9 years agomalta: disable the openwrt commandline hack
Jonas Gorski [Sun, 30 Aug 2015 12:18:05 +0000 (12:18 +0000)]
malta: disable the openwrt commandline hack

We don't make use of it, so no need to have it enabled.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46758

9 years agobrcm63xx: disable the openwrt commandline hack
Jonas Gorski [Sun, 30 Aug 2015 12:18:00 +0000 (12:18 +0000)]
brcm63xx: disable the openwrt commandline hack

We don't make use of it, so no need to have it enabled.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46757

9 years agobrcm47xx: disable the openwrt commandline hack
Jonas Gorski [Sun, 30 Aug 2015 12:17:57 +0000 (12:17 +0000)]
brcm47xx: disable the openwrt commandline hack

We don't make use of it, so no need to have it enabled.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46756

9 years agoath25: disable the openwrt commandline hack
Jonas Gorski [Sun, 30 Aug 2015 12:17:56 +0000 (12:17 +0000)]
ath25: disable the openwrt commandline hack

We don't make use of it, so no need to have it enabled.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46755

9 years agoadm5120: disable the openwrt commandline hack
Jonas Gorski [Sun, 30 Aug 2015 12:17:54 +0000 (12:17 +0000)]
adm5120: disable the openwrt commandline hack

We don't make use of it, so no need to have it enabled.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46754

9 years agobrcm63xx: remove legacy led/button related patches
Jonas Gorski [Sun, 30 Aug 2015 12:17:52 +0000 (12:17 +0000)]
brcm63xx: remove legacy led/button related patches

We register all gpio buttons and leds through DT, so no need to keep
fixes/additions for the platform data based bay.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46753

9 years agobrcm63xx: add Comtrend VR-3026e support
Jonas Gorski [Sun, 30 Aug 2015 12:17:46 +0000 (12:17 +0000)]
brcm63xx: add Comtrend VR-3026e support

Add support for Comtrend VR-3026e v1.
The device is almost identical to the Comtrend VR-3025un.

Signed-off-by: Martin Tesar <tesarmar@gmail.com>
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46752

9 years agobroadcom-wl: add interface after bridge is up
Hauke Mehrtens [Sat, 29 Aug 2015 20:05:04 +0000 (20:05 +0000)]
broadcom-wl: add interface after bridge is up

Without this patch adding the wifi device to the bridge may fail
because the bridge is not already configured when the wifi device gets
configured. This patch makes broadcom-wl wait till the bridge is ready.

This fixes #17262

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46749

9 years agobroadcom-wl: fix bash comparison error
Hauke Mehrtens [Sat, 29 Aug 2015 17:38:15 +0000 (17:38 +0000)]
broadcom-wl: fix bash comparison error

This fixes #17495

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46746

9 years agobroadcom-wl: fix section mismatches in broadcom-wl driver
Hauke Mehrtens [Sat, 29 Aug 2015 17:37:41 +0000 (17:37 +0000)]
broadcom-wl: fix section mismatches in broadcom-wl driver

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46745

9 years agobroadcom-wl: fix build with kernel 4.1
Hauke Mehrtens [Sat, 29 Aug 2015 17:37:09 +0000 (17:37 +0000)]
broadcom-wl: fix build with kernel 4.1

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46744

9 years agobrcm47xx: fix Linksys E1200 V2 image filename
Rafał Miłecki [Sat, 29 Aug 2015 09:23:19 +0000 (09:23 +0000)]
brcm47xx: fix Linksys E1200 V2 image filename

Drop unwanted suffix (copy & paste mistake).

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46742

9 years agomt76: update to the latest version, fixes some tx power issues
Felix Fietkau [Fri, 28 Aug 2015 07:05:41 +0000 (07:05 +0000)]
mt76: update to the latest version, fixes some tx power issues

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 46741

9 years agomac80211: add pending brcmfmac patch fixing warm reboots
Rafał Miłecki [Thu, 27 Aug 2015 14:39:30 +0000 (14:39 +0000)]
mac80211: add pending brcmfmac patch fixing warm reboots

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46739

9 years agouci: update to the latest version, adds various fixes
Felix Fietkau [Thu, 27 Aug 2015 12:43:27 +0000 (12:43 +0000)]
uci: update to the latest version, adds various fixes

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 46738

9 years agohostapd: properly enable 802.11w support
Felix Fietkau [Thu, 27 Aug 2015 12:43:22 +0000 (12:43 +0000)]
hostapd: properly enable 802.11w support

Add CONFIG_IEEE80211W variable to DRIVER_MAKEOPTS so that 802.11w
support is properly compiled in full variant.

This fixes #20179

Signed-off-by: Janusz Dziemidowicz <rraptorr@nails.eu.org>
SVN-Revision: 46737

9 years agomac80211: add pending brcmfmac patches fixing multiple interfaces
Rafał Miłecki [Wed, 26 Aug 2015 22:10:14 +0000 (22:10 +0000)]
mac80211: add pending brcmfmac patches fixing multiple interfaces

So far support for multiple interface was somehow broken in brcmfmac.
Driver couldn't correctly match firmware and system interfaces resulting
in not working APs and WARNINGs. This pending patches fixes that :)

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46734

9 years agobcm53xx: support sysupgrade with Netgear R7000 original firmware
Rafał Miłecki [Wed, 26 Aug 2015 15:21:14 +0000 (15:21 +0000)]
bcm53xx: support sysupgrade with Netgear R7000 original firmware

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46732

9 years agobcm53xx: add OpenWrt specific stuff for Netgear R7000
Rafał Miłecki [Wed, 26 Aug 2015 15:21:07 +0000 (15:21 +0000)]
bcm53xx: add OpenWrt specific stuff for Netgear R7000

This adds DT things that couldn't be upstreamed yet.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46731

9 years agobcm53xx: use pending Netgear R7000 patch
Rafał Miłecki [Wed, 26 Aug 2015 14:51:47 +0000 (14:51 +0000)]
bcm53xx: use pending Netgear R7000 patch

It includes support for LEDs and buttons.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46730

9 years agobcm53xx: use backported patches for UART0 and profiling
Rafał Miłecki [Wed, 26 Aug 2015 14:39:06 +0000 (14:39 +0000)]
bcm53xx: use backported patches for UART0 and profiling

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46729

9 years agobrcm63xx: fix WAP-5813n default network config
Jonas Gorski [Wed, 26 Aug 2015 13:03:09 +0000 (13:03 +0000)]
brcm63xx: fix WAP-5813n default network config

/etc/uci-defaults/02_network had a typo, making it generate the wrong
network config.

Closes #20407.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46727

9 years agolinux: fix off-by-one in handling in /proc/net/route
Jonas Gorski [Wed, 26 Aug 2015 10:11:41 +0000 (10:11 +0000)]
linux: fix off-by-one in handling in /proc/net/route

Add an upstream fix for /proc/net/route causing missing routes doing
several continued reads from it.

Only 4.1+ is affected.

Closes #20403.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46726

9 years agobrcm63xx: enable dual rx/tx spi support for hsspi
Jonas Gorski [Tue, 25 Aug 2015 22:04:53 +0000 (22:04 +0000)]
brcm63xx: enable dual rx/tx spi support for hsspi

should improve flash access times. Should be harmless to gnerally
enable regardless if a flash supporting dual reads is attached. In
doubt, spi-nor will just fall back to serial reads.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46725

9 years agomac80211: backport latest patches except for NVRAM support
Rafał Miłecki [Tue, 25 Aug 2015 15:20:48 +0000 (15:20 +0000)]
mac80211: backport latest patches except for NVRAM support

We sill don't use kernel 4.2 which is required for backporting using
upstream NVRAM support patch.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46724

9 years agoscripts/patch-kernel.sh: remove -E flag to preserve empty files touched by patches
Felix Fietkau [Tue, 25 Aug 2015 07:46:32 +0000 (07:46 +0000)]
scripts/patch-kernel.sh: remove -E flag to preserve empty files touched by patches

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 46723

9 years agorules.mk: remove GCC 4.4 and GCC 4.5 conditional
Felix Fietkau [Tue, 25 Aug 2015 07:45:42 +0000 (07:45 +0000)]
rules.mk: remove GCC 4.4 and GCC 4.5 conditional

gcc 4.4 was removed in r44957 gcc: remove 4.4.7 (only used by avr32)
gcc 4.5 was removed in r36149

Signed-off-by: Dirk Neukirchen <dirkneukirchen@web.de>
SVN-Revision: 46722

9 years agokernel: bridge, multicast-to-unicast: assign src after pskb_may_pull()
Felix Fietkau [Tue, 25 Aug 2015 07:25:20 +0000 (07:25 +0000)]
kernel: bridge, multicast-to-unicast: assign src after pskb_may_pull()

A call to pskb_may_pull() might reallocate skb->data. Therefore we
should only assign the src-pointer after any potential reallocations.

Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 46721

9 years agoRevert "kernel: disable multicast-to-unicast translation for ipv6 neighbor solicitati...
Felix Fietkau [Tue, 25 Aug 2015 07:25:10 +0000 (07:25 +0000)]
Revert "kernel: disable multicast-to-unicast translation for ipv6 neighbor solicitation (#17625)"

This reverts commit a080e8e1943156168913d0353a2e99d1151102aa.

It did not fix the problem but just hid some symptom. The real issue was
that IGMP/MLD report suppression was not considered for the
multicast-to-unicast feature. A recent netifd which isolates IGMP/MLD
reports between STAs by utilizing AP-isolation and bridge-hairpinning
should have fixed this.

It is perfectly fine to apply multicast-to-unicast to IPv6 Neighbor
Solicitations, too (once that feature is configured correctly).

Signed-off-by: Linus Lüssing <linus.luessing@c0d3.blue>
Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 46720

9 years agonetifd: update to the latest version, adds multicast-to-unicast fixes
Felix Fietkau [Tue, 25 Aug 2015 07:24:53 +0000 (07:24 +0000)]
netifd: update to the latest version, adds multicast-to-unicast fixes

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 46719

9 years agonetifd: various updates
Steven Barth [Tue, 25 Aug 2015 06:27:37 +0000 (06:27 +0000)]
netifd: various updates

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 46718

9 years agoomcproxy: fix last commit
Steven Barth [Mon, 24 Aug 2015 08:53:16 +0000 (08:53 +0000)]
omcproxy: fix last commit

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 46717

9 years agoomcproxy: add new igmpv3 & mldv2 multicast proxy
Steven Barth [Mon, 24 Aug 2015 08:28:11 +0000 (08:28 +0000)]
omcproxy: add new igmpv3 & mldv2 multicast proxy

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 46716

9 years agokernel: remove already upstream applied patch
Jonas Gorski [Sun, 23 Aug 2015 21:33:57 +0000 (21:33 +0000)]
kernel: remove already upstream applied patch

Remove 131-MIPS-export-get_c0_perfcount_int.patch which was already applied
in 4.1.6. This fixes the following build error:

arch/mips/ath79/setup.c:217:77: error: redefinition of '__kstrtab_get_c0_perfcount_int'
arch/mips/ath79/setup.c:211:77: note: previous definition of '__kstrtab_get_c0_perfcount_int' was here
arch/mips/ath79/setup.c:217:350: error: redefinition of '__ksymtab_get_c0_perfcount_int'
arch/mips/ath79/setup.c:211:350: note: previous definition of '__ksymtab_get_c0_perfcount_int' was here
scripts/Makefile.build:258: recipe for target 'arch/mips/ath79/setup.o' failed

Reported-by: swalker
Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46715

9 years agoadm5120: make patches apply again
Jonas Gorski [Sun, 23 Aug 2015 19:11:07 +0000 (19:11 +0000)]
adm5120: make patches apply again

Fix patches not applying anymore since r46654.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46714

9 years agokernel: update 4.1 to 4.1.6
Jonas Gorski [Sun, 23 Aug 2015 18:06:11 +0000 (18:06 +0000)]
kernel: update 4.1 to 4.1.6

Changelog:
* https://www.kernel.org/pub/linux/kernel/v4.x/ChangeLog-4.1.6

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46713

9 years agomvebu: kirkwood: fix ehci-orion probe if generic-phy isn't enabled
Jonas Gorski [Sun, 23 Aug 2015 13:35:03 +0000 (13:35 +0000)]
mvebu: kirkwood: fix ehci-orion probe if generic-phy isn't enabled

Properly treat -ENOSYS as no PHY, else ehci-orion won't work without
generic phy support.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46711

9 years agomac80211: disable ipw2x00 for big endian targets
Jonas Gorski [Sun, 23 Aug 2015 09:36:03 +0000 (09:36 +0000)]
mac80211: disable ipw2x00 for big endian targets

The ipw2x00 drivers assume that the system they are running is little
endian, and access everything in native byte order. When run on a big
endian system, everything breaks apart.

Since fixing this is non trivial on a first glance, disable them for
big endian targets.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46708

9 years agobrcm63xx: ensure dummy byte is set for mapped spi flash with fast read
Jonas Gorski [Sun, 23 Aug 2015 09:35:56 +0000 (09:35 +0000)]
brcm63xx: ensure dummy byte is set for mapped spi flash with fast read

Some CFEs seem to misconfigure the mapped memory flash access with
fast read but without a dummy byte, causing all accesses to be prefixed
with 0xff.
This of course breaks reading out the nvram, so do not just move back to
single i/o accessors, but also ensure that the dummy byte is correctly
set.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46707

9 years agoscripts/feeds: allow export of feeds with static revisions
Jonas Gorski [Sun, 23 Aug 2015 09:35:46 +0000 (09:35 +0000)]
scripts/feeds: allow export of feeds with static revisions

Add a switch to scripts/feeds that allows listing the feeds with their
currently checked out revisions in feeds.conf compatible format.

This allows providing a feeds.conf for public builds to make replication
of the build easier.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46706

9 years agosunxi: bump to 4.1
Zoltan Herpai [Fri, 21 Aug 2015 13:23:29 +0000 (13:23 +0000)]
sunxi: bump to 4.1

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
SVN-Revision: 46705

9 years agosunxi: add audio module for 4.1
Zoltan Herpai [Fri, 21 Aug 2015 13:16:11 +0000 (13:16 +0000)]
sunxi: add audio module for 4.1

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
SVN-Revision: 46704

9 years agosunxi: update ahci package list for a couple board profiles
Zoltan Herpai [Fri, 21 Aug 2015 10:16:15 +0000 (10:16 +0000)]
sunxi: update ahci package list for a couple board profiles

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
SVN-Revision: 46703

9 years agosunxi: reorganize kmod-ata-sunxi dependencies
Zoltan Herpai [Fri, 21 Aug 2015 10:08:07 +0000 (10:08 +0000)]
sunxi: reorganize kmod-ata-sunxi dependencies

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 46702

9 years agoNewer kernels make use of libahci in ahci-platform, ie. also on non-PCI platforms...
Zoltan Herpai [Fri, 21 Aug 2015 09:48:05 +0000 (09:48 +0000)]
Newer kernels make use of libahci in ahci-platform, ie. also on non-PCI platforms. As there is no single config symbol deciding whether libahci.ko is being built or not it seems the most sensitive thing to package it in kmod-ata-core.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 46701

9 years agoThis had been set in r44508 as a workaround for switch problems.
Imre Kaloz [Fri, 21 Aug 2015 08:10:48 +0000 (08:10 +0000)]
This had been set in r44508 as a workaround for switch problems.

Now that the switch driver can handle two devices with
the same MAC address in separate VLANs we can go back
to using the same address on both interfaces.

This is the Linksys firmware's default behavior.

Signed-off-by: Claudio Leite <leitec@staticky.com>
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 46700

9 years agoPreviously, all VLANs (port-based or 802.1q) were sharing a single database in the...
Imre Kaloz [Fri, 21 Aug 2015 08:09:52 +0000 (08:09 +0000)]
Previously, all VLANs (port-based or 802.1q) were sharing a single database in the ATU. This created problems in the case of a system where two ports/devices share a MAC address (e.g. Linksys WRT1900AC eth0/eth1).

This also clears any bootloader-set FDB defaults. This had
caused issues creating port-based VLANs when mappings
overlapped previous VLANs. Packets destined to a port
not in the default port group flooded all ports.

Tested on a 88E6171 (Linksys EA4500) and 88E6172 ('1900AC)

Signed-off-by: Claudio Leite <leitec@staticky.com>
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 46699

9 years agomvebu: fix cobra chainmask
Imre Kaloz [Fri, 21 Aug 2015 08:01:48 +0000 (08:01 +0000)]
mvebu: fix cobra chainmask

Thanks Sebastian for spotting it.

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 46697

9 years agokernel: add missing symbols for 4.1
Zoltan Herpai [Thu, 20 Aug 2015 23:19:42 +0000 (23:19 +0000)]
kernel: add missing symbols for 4.1

Signed-off-by: Zoltan HERPAI <wigyori@uid0.hu>
SVN-Revision: 46696

9 years agoodhcpd: various RA improvements
Steven Barth [Thu, 20 Aug 2015 12:43:45 +0000 (12:43 +0000)]
odhcpd: various RA improvements

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 46694

9 years agomwlwifi: fix commit hash for 10.3.0.8
Imre Kaloz [Wed, 19 Aug 2015 08:45:19 +0000 (08:45 +0000)]
mwlwifi: fix commit hash for 10.3.0.8

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 46693

9 years agomvebu: add support for the Linksys boot counter
Imre Kaloz [Wed, 19 Aug 2015 08:31:23 +0000 (08:31 +0000)]
mvebu: add support for the Linksys boot counter

The u-boot boot counter was never reset after a successful boot,
which sometimes could make some variables become out of sync.
This patch adds support for the boot counter and enables
auto_recovery unconditionally.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Signed-off-by: Rob Mosher <nyt-openwrt@countercultured.net>
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 46690

9 years agouhttpd: we don't know where the device is located, so reflect that in the cert
Imre Kaloz [Wed, 19 Aug 2015 08:20:11 +0000 (08:20 +0000)]
uhttpd: we don't know where the device is located, so reflect that in the cert

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 46688

9 years agomwlwifi: upgrade to 10.3.0.8-20150818
Imre Kaloz [Tue, 18 Aug 2015 11:22:47 +0000 (11:22 +0000)]
mwlwifi: upgrade to 10.3.0.8-20150818

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 46687

9 years agoenable strong SSP / Stackprotector on gcc5
Steven Barth [Tue, 18 Aug 2015 09:20:34 +0000 (09:20 +0000)]
enable strong SSP / Stackprotector on gcc5

Signed-off-by: Dirk Neukirchen <dirkneukirchen@web.de>
SVN-Revision: 46685

9 years agomusl: fix getsubopt function
Steven Barth [Tue, 18 Aug 2015 09:18:40 +0000 (09:18 +0000)]
musl: fix getsubopt function

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 46684

9 years agopolarssl: Fix build failures due to PKG_NAME != dir name
Steven Barth [Tue, 18 Aug 2015 08:37:38 +0000 (08:37 +0000)]
polarssl: Fix build failures due to PKG_NAME != dir name

Packages that depend on PolarSSL fail to build because polarssl's InstallDev
section never actually gets executed because (prior to this patch) the package
name does not match the subdir the package is in (presumably due to upstream
name change).  As a workaround I have changed the package name back to
polarssl and used a new variable SRC_PKG_NAME for the purposes of downloading
the upstream tarball and creating PKG_BUILD_DIR.

Signed-off-by: Daniel Dickinson <openwrt@daniel.thecshore.com>
SVN-Revision: 46683

9 years agomac80211: add mt7601u driver
Hauke Mehrtens [Mon, 17 Aug 2015 21:19:27 +0000 (21:19 +0000)]
mac80211: add mt7601u driver

firmware is not available in linux-firmware yet
so use official driver source

Signed-off-by: Dirk Neukirchen <dirkneukirchen@web.de>
SVN-Revision: 46682

9 years agouhttpd: pass X-HTTP-Method-Override header to cgi scripts
Jo-Philipp Wich [Mon, 17 Aug 2015 16:17:36 +0000 (16:17 +0000)]
uhttpd: pass X-HTTP-Method-Override header to cgi scripts

Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
SVN-Revision: 46677

9 years agoat91: fix usb rate backport patch tab mangle
John Crispin [Mon, 17 Aug 2015 16:10:49 +0000 (16:10 +0000)]
at91: fix usb rate backport patch tab mangle

The patch submitted in [46649] was mangled in the use of gmails webmail interface, tabs replaced with spaces, resulting in a patch which dit not apply.
This should fix the issue, sorry for the noise.

Signed-off-by: Ben Whitten <benwhitten@gmail.com>
SVN-Revision: 46676

9 years agoorion: remove IRQF_DISABLED
John Crispin [Mon, 17 Aug 2015 11:26:14 +0000 (11:26 +0000)]
orion: remove IRQF_DISABLED

no-op since 2.6.35
    removed in Kernel 4.1
    see https://lwn.net/Articles/380931/

Signed-off-by: Dirk Neukirchen <dirkneukirchen@web.de>
SVN-Revision: 46672

9 years agogeneric: crypto: remove IRQF_DISABLED
John Crispin [Mon, 17 Aug 2015 11:26:02 +0000 (11:26 +0000)]
generic: crypto: remove IRQF_DISABLED

no-op since 2.6.35
    removed in Kernel 4.1
    see https://lwn.net/Articles/380931/

Signed-off-by: Dirk Neukirchen <dirkneukirchen@web.de>
SVN-Revision: 46671

9 years agolantiq: Support newer versions of the PEF7071 ethernet PHYs
John Crispin [Mon, 17 Aug 2015 11:25:57 +0000 (11:25 +0000)]
lantiq: Support newer versions of the PEF7071 ethernet PHYs

The BT Home Hub 5A uses three PEF7071 with PHY ID 0xd565a401. Daniel's
PHY driver (for his u-boot sources) already supports that PHY because
it uses a PHY ID mask of 0xfffffff8.

Signed-off-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com>
SVN-Revision: 46670

9 years agoiwinfo: update to latest git HEAD
John Crispin [Mon, 17 Aug 2015 11:25:51 +0000 (11:25 +0000)]
iwinfo: update to latest git HEAD

adds extra station info reporting

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 46669

9 years agooxnas: remove support for kernel 3.18 and 4.0
John Crispin [Mon, 17 Aug 2015 11:25:46 +0000 (11:25 +0000)]
oxnas: remove support for kernel 3.18 and 4.0

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 46668

9 years agoramips: move patches around
John Crispin [Mon, 17 Aug 2015 10:13:46 +0000 (10:13 +0000)]
ramips: move patches around

0073 needs to applied afert 0102

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 46667

9 years agougps: update to latest git HEAD
John Crispin [Mon, 17 Aug 2015 10:13:03 +0000 (10:13 +0000)]
ugps: update to latest git HEAD

fixes 2 bugs
* typo
* use %04d rather than %d when printing the position

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 46666

9 years agoramips: base-files: optimize etc/diag.sh and etc/board.d/01_leds scripts
John Crispin [Mon, 17 Aug 2015 06:24:47 +0000 (06:24 +0000)]
ramips: base-files: optimize etc/diag.sh and etc/board.d/01_leds scripts

Instead of using board name provided explicitly as text in LED name, ex.:

[...]
mlwg2)
status_led="mlwg2:blue:system"
[...]

use $board variable, which allows to combine together multiple boards with same color and LED names, ex:

[...]
mlw221|\
mlwg2)
status_led="$board:blue:system"
[...]

The above approach allows to shrink size of code in base-files/etc/board.d/01_leds and base-files/etc/diag.sh scripts dramatically.
One thing to keep in mind here is that we assume to use proper and consistent LED naming scheme ("device:color:led-name").

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
SVN-Revision: 46665

9 years agoramips: use consistent naming scheme for LEDs
John Crispin [Mon, 17 Aug 2015 06:24:34 +0000 (06:24 +0000)]
ramips: use consistent naming scheme for LEDs

The upstream LED naming convention is "device:color:led-name", but it seems that many of supported boards in OpenWrt don't follow this approach.
The following patch fixes this inconsistency in dts{,i} files and updates base-files scripts for ramips target:
 * fixes wrong indentation
 * keeps case statements structure in same convention as in other scripts (no empty line after ";;", no indentation for case...esac body)
 * fixes wrong LED names for some of boards (makes them the same as in dts{,i} files)
 * combines boards with same configuration (ex. set_wifi_led "rt2800pci-phy0::radio" in 01_leds)

Signed-off-by: Piotr Dymacz <pepe2k@gmail.com>
SVN-Revision: 46664

9 years agoar71xx: add support for TP-Link TL-WDR6500 v2
John Crispin [Mon, 17 Aug 2015 06:24:04 +0000 (06:24 +0000)]
ar71xx: add support for TP-Link TL-WDR6500 v2

Signed-off-by: Weijie Gao <hackpascal@gmail.com>
This patch adds support for TP-Link TL-WDR6500 v2.

The firmware has a U-Boot header for kernel, and a TP-LINK v2 header for
the whole firmware, so I have to create a new firmware creation method.

SVN-Revision: 46663

9 years agoar71xx: add 64kb bootloader mtd parser for tplinkpart
John Crispin [Mon, 17 Aug 2015 06:23:49 +0000 (06:23 +0000)]
ar71xx: add 64kb bootloader mtd parser for tplinkpart

Signed-off-by: Weijie Gao <hackpascal@gmail.com>
This patch adds flash layout parser for TP-Link firmwares which have a 64kb
bootloader.

This is used for TP-Link TL-WDR6500 v2.

SVN-Revision: 46662

9 years agotools/firmware-utils: add header version 2 support for mktplinkfw
John Crispin [Mon, 17 Aug 2015 06:23:43 +0000 (06:23 +0000)]
tools/firmware-utils: add header version 2 support for mktplinkfw

Signed-off-by: Weijie Gao <hackpascal@gmail.com>
This patch adds header version 2 option for mktplinkfw.

The version 2 header is used for AR/QCA firmwares and is not the same as
the header generated by mktplinkfw2.
Instead, it is nearly the same as version 1 header except for the header
version and the RSA signature.

The header version 2 support is used for newer TP-Link routers which have
only a 64kb bootloader part, e.g. TL-WDR6500 v2.

SVN-Revision: 46661

9 years agoramips: reorder the sysupgrade script
John Crispin [Mon, 17 Aug 2015 06:18:38 +0000 (06:18 +0000)]
ramips: reorder the sysupgrade script

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 46660

9 years agoramips: reorder patches
John Crispin [Mon, 17 Aug 2015 06:18:30 +0000 (06:18 +0000)]
ramips: reorder patches

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 46659

9 years agoipq806x: switch AP148 to using SMEM based MTD parser
John Crispin [Mon, 17 Aug 2015 06:18:15 +0000 (06:18 +0000)]
ipq806x: switch AP148 to using SMEM based MTD parser

*Enable SMEM MTD parser and its dependencies (SMEM & HW spinlocks) in
 the kernel config
*Replaces the MTD layout in DT by the dynamic layout provided by the
 SMEM parser for AP148

Using the OF based parser is still possible on platforms which have a
fixed MTD partition layout.

Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 46658

9 years agoipq806x: add QCOM SMEM based MTD parser
John Crispin [Mon, 17 Aug 2015 06:18:04 +0000 (06:18 +0000)]
ipq806x: add QCOM SMEM based MTD parser

This patch adds a new parser which uses the SMEM available on IPQ and
some other QCOM platforms to map the MTD partitions.

Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 46657

9 years agoipq806x: add SMEM support
John Crispin [Mon, 17 Aug 2015 06:17:58 +0000 (06:17 +0000)]
ipq806x: add SMEM support

2 patches are cherry-picked from the following LKML thread:
*https://lkml.org/lkml/2015/4/11/208

The last patch (036-soc-qcom-add-smem-to-IPQ806x-platforms.patch) is
adding the corresponding DT nodes required for IPQ806x.

Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 46656

9 years agoipq806x: add hwspinlock support
John Crispin [Mon, 17 Aug 2015 06:17:47 +0000 (06:17 +0000)]
ipq806x: add hwspinlock support

This change cherry-picks the following 3 changes from linux-next:
*fb7737 hwspinlock/core: add device tree support
*19a0f6 hwspinlock: qcom: Add support for Qualcomm HW Mutex block
*bd5717 hwspinlock: qcom: Correct msb in regmap_field

We're also adding a patch to add the hardware spinlock device nodes on
IPQ806x platforms (033-soc-qcom-Add-sfbp-device-to-IPQ806x-dts.patch).

Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 46655

9 years agokernel: move linux, part-probe dt parsing in generic code
John Crispin [Mon, 17 Aug 2015 06:17:37 +0000 (06:17 +0000)]
kernel: move linux, part-probe dt parsing in generic code

The "linux,part-probe" dts parsing is a pretty neat generic feature.
It has been posted to kernel.org and could easily be reused by all
targets.

This change moves the patch to the 3.18 and 4.1 generic folders, and
makes the feature available to all platforms who may want to use it.

Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
Acked-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46654

9 years agobase-files: default_postinst: propagate the real postinst return code
John Crispin [Mon, 17 Aug 2015 06:17:23 +0000 (06:17 +0000)]
base-files: default_postinst: propagate the real postinst return code

Using the postinst script for sanity checks and expecting opkg to fail
if the postinst didn't return 0 was possible in Barrier Breaker, propagate
the real postinst return code through default_postinst to restore this
behaviour.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
SVN-Revision: 46653

9 years agohostapd: Add vlan_file option to netifd.sh
John Crispin [Mon, 17 Aug 2015 06:17:13 +0000 (06:17 +0000)]
hostapd: Add vlan_file option to netifd.sh

Other VLAN related options are already being processed in netifd.sh
but the vlan_file option is missing. This option allows the mapping
of vlan IDs to network interfaces and will be used in dynamic VLAN
feature for binding stations to interfaces based on VLAN
assignments. The change is done similarly to the wpa_psk_file
option.

Signed-off-by: Gong Cheng <chengg11@yahoo.com>
SVN-Revision: 46652

9 years agoar71xx: Add support for Black Swift
John Crispin [Mon, 17 Aug 2015 06:16:54 +0000 (06:16 +0000)]
ar71xx: Add support for Black Swift

Add support for the Black Swift board:
http://www.black-swift.com/

Took relevant parts and slightly adapted from:
https://github.com/blackswift/openwrt/blob/master/bsb.patch

Signed-off-by: Bruno Randolf <br1@einfach.org>
SVN-Revision: 46651

9 years agobase-files: rework reset button script to indicate factory reset
John Crispin [Mon, 17 Aug 2015 06:16:39 +0000 (06:16 +0000)]
base-files: rework reset button script to indicate factory reset

With this change we let user know 5 seconds have passed and reset button
can be released (to trigger factory reset)

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 46650

9 years agoat91: Backport determine_rate prototype fix
John Crispin [Mon, 17 Aug 2015 06:16:29 +0000 (06:16 +0000)]
at91: Backport determine_rate prototype fix

This patch is backport of a fix to a USB prototype mismatch bug found
in 3.18 kernels, originally submitted by Boris Brezillon [1].
The symptom of this bug was that devices attached to the at91 using
at91_ohci on a hub never appeared and failed to initialise.

[1] http://www.spinics.net/lists/linux-usb/msg125969.html

Signed-off-by: Ben Whitten <ben.whitten@gmail.com>
SVN-Revision: 46649

9 years agomcs814x: use firmware partition splitter on dLAN USB Extender
John Crispin [Mon, 17 Aug 2015 06:16:19 +0000 (06:16 +0000)]
mcs814x: use firmware partition splitter on dLAN USB Extender

to avoid editing the dts every time the kernel size changes.
uImage is now bigger than 1MB. Pad uImage to 64k erase block size.

Signed-off-by: Günther Kelleter <guenther.kelleter@devolo.de>
SVN-Revision: 46648

9 years agomcs814x: fix interrupt handling
John Crispin [Mon, 17 Aug 2015 06:16:08 +0000 (06:16 +0000)]
mcs814x: fix interrupt handling

Switch to generich chip irqs/irq domains.
Interrupts were broken since kernel 3.14. dLAN USB extender is now
booting again.

Signed-off-by: Günther Kelleter <guenther.kelleter@devolo.de>
SVN-Revision: 46647

9 years agomcs814x: fix debug UART
John Crispin [Mon, 17 Aug 2015 06:15:56 +0000 (06:15 +0000)]
mcs814x: fix debug UART

Switch to new 8250 debug uart code because the old
mach-mcs814x/include/mach/debug-macro.S tries to include
asm/hardware/debug-8250.S which no longer exists since kernel 3.14

Signed-off-by: Günther Kelleter <guenther.kelleter@devolo.de>
SVN-Revision: 46646

9 years agoramips: Fix MT7621 pinmux bits for uart3, uart2, mdio
John Crispin [Mon, 17 Aug 2015 06:15:49 +0000 (06:15 +0000)]
ramips: Fix MT7621 pinmux bits for uart3, uart2, mdio

The MT7621 uses a 2 bit wide configuration of the sdhci, spi, mdio, pcie,
wdt, uart2 and uart3 in the GPIO_MODE register. It was correctly done
for sdhci, spi, pcie and wdt, The same has to be done for uart3, uart2
and mdio.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 46645

9 years agoramips: Fix amount of MT7621 pins controlled by spi group
John Crispin [Mon, 17 Aug 2015 06:15:44 +0000 (06:15 +0000)]
ramips: Fix amount of MT7621 pins controlled by spi group

The PINS conntrolled by the SPI bits in the GPIO_MODE register is always
7 and not 8 for nand mode.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 46644

9 years agoramips: Fix uart2/uart3 pinmux order on MT7621
John Crispin [Mon, 17 Aug 2015 06:15:34 +0000 (06:15 +0000)]
ramips: Fix uart2/uart3 pinmux order on MT7621

The uart3 setting in GPIO_MODE register is before the uart2 setting. Also
don't mix uart2 and uart3 function/groups.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
SVN-Revision: 46643

9 years agobrcm2708: override disabled device tree module status from config.txt
John Crispin [Mon, 17 Aug 2015 06:04:48 +0000 (06:04 +0000)]
brcm2708: override disabled device tree module status from config.txt

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 46642

9 years agobrcm2708: fix kmod-sound-soc-iqaudio-dac and add missing symbol
John Crispin [Mon, 17 Aug 2015 06:04:39 +0000 (06:04 +0000)]
brcm2708: fix kmod-sound-soc-iqaudio-dac and add missing symbol

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 46641

9 years agobrcm2708: update 4.1 patches
John Crispin [Mon, 17 Aug 2015 06:04:32 +0000 (06:04 +0000)]
brcm2708: update 4.1 patches

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 46640