openwrt/staging/jow.git
9 years agoatheros: v3.18: cleanup register headers
Felix Fietkau [Fri, 13 Mar 2015 03:00:59 +0000 (03:00 +0000)]
atheros: v3.18: cleanup register headers

AFAIK, no one AR2315+ chip (AR2315, AR2316, AR2317, AR2318) does not
contain IR block, so remove IR registers definitions. Also remove few
unused macroses.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44722

9 years agoatheros: v3.18: pass PCI IRQ and I/O mem via resources
Felix Fietkau [Fri, 13 Mar 2015 03:00:51 +0000 (03:00 +0000)]
atheros: v3.18: pass PCI IRQ and I/O mem via resources

Pass PCI IRQ and I/O memory ranges via platform device resources, this
change makes PCI controller driver independed from arch headers, so
also remove few includes.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44721

9 years agoatheros: v3.18: remap PCI controller MMR memory
Felix Fietkau [Fri, 13 Mar 2015 03:00:43 +0000 (03:00 +0000)]
atheros: v3.18: remap PCI controller MMR memory

Honestly remap PCI controller MMR and use accessor functions to interact
with registers.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44720

9 years agoatheros: v3.18: add context container for PCI driver
Felix Fietkau [Fri, 13 Mar 2015 03:00:19 +0000 (03:00 +0000)]
atheros: v3.18: add context container for PCI driver

Add container and place all context specific variables and structure to
it.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44719

9 years agoatheros: v3.18: move PCI enable code to arch
Felix Fietkau [Fri, 13 Mar 2015 03:00:06 +0000 (03:00 +0000)]
atheros: v3.18: move PCI enable code to arch

Move PCI host interface enable code to arch, since it touches generic
SoC registers outside the PCI MMR region.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44718

9 years agoatheros: v3.18: rearrange PCI regs definitions
Felix Fietkau [Fri, 13 Mar 2015 02:59:54 +0000 (02:59 +0000)]
atheros: v3.18: rearrange PCI regs definitions

Move PCI controller configuration registers from generic header to
driver source. No functional changes.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44717

9 years agoatheros: v3.18: relocate PCI host DMA base definition
Felix Fietkau [Fri, 13 Mar 2015 02:59:43 +0000 (02:59 +0000)]
atheros: v3.18: relocate PCI host DMA base definition

Put AR2315_PCI_HOST_SDRAM_BASEADDR macro to DMA header, since this is
arbitrary value and not some hw specific constant. Also this relocation
decouples dma from HW specific header.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44716

9 years agoatheros: v3.18: rework early initialization
Felix Fietkau [Fri, 13 Mar 2015 02:59:27 +0000 (02:59 +0000)]
atheros: v3.18: rework early initialization

Do not use prom_init() callback, do memory initialization in
plat_mem_setup() callback and move serial port configuration to
arch_initcall stage.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44715

9 years agoatheros: v3.18: rearrange interrupt handling functions
Felix Fietkau [Fri, 13 Mar 2015 02:59:19 +0000 (02:59 +0000)]
atheros: v3.18: rearrange interrupt handling functions

No functional changes, just change functions order in source file.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44714

9 years agoatheros: v3.18: change calls logic
Felix Fietkau [Fri, 13 Mar 2015 02:59:04 +0000 (02:59 +0000)]
atheros: v3.18: change calls logic

Check SoC family (AR5312+ or AR2315+) before call instead of checking it
inside the called function. Also convert ar{5312,2315}_init_device()
function to void, since they both return zero and nobody care about
return value.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44713

9 years agoatheros: v3.18: update names and declarations
Felix Fietkau [Fri, 13 Mar 2015 02:58:58 +0000 (02:58 +0000)]
atheros: v3.18: update names and declarations

Sync functions, variables and enums names with upstream. Mostly replace
'ar231x_' prefix by 'ath25_'.

No functional changes, except few 'int' -> 'unsigned' changes.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44712

9 years agoatheros: v3.18: cleanup cpu-feature-overrides.h
Felix Fietkau [Fri, 13 Mar 2015 02:58:49 +0000 (02:58 +0000)]
atheros: v3.18: cleanup cpu-feature-overrides.h

Remove mention of undefined features.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44711

9 years agoatheros: v3.18: remove eth platform device unused field
Felix Fietkau [Fri, 13 Mar 2015 02:58:37 +0000 (02:58 +0000)]
atheros: v3.18: remove eth platform device unused field

Ethernet controller driver don't use boarddata directly, so remove
corresponding field from its platform device structure.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44710

9 years agoatheros: v3.18: remove odd header reset.h
Felix Fietkau [Fri, 13 Mar 2015 02:58:24 +0000 (02:58 +0000)]
atheros: v3.18: remove odd header reset.h

This header provides prototype for function without realization and
users.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44709

9 years agoatheros: v3.18: rename platform header
Felix Fietkau [Fri, 13 Mar 2015 02:58:17 +0000 (02:58 +0000)]
atheros: v3.18: rename platform header

Rename ar231x_platform.h to ath25_platform.h

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44708

9 years agoatheros: v3.18: rename kernel platform to ath25
Felix Fietkau [Fri, 13 Mar 2015 02:57:59 +0000 (02:57 +0000)]
atheros: v3.18: rename kernel platform to ath25

Rename platform source directory and asm includes directory.

No functional changes.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44707

9 years agoatheros: v3.18: rename ATHEROS_AR231X symbol to ATH25
Felix Fietkau [Fri, 13 Mar 2015 02:57:51 +0000 (02:57 +0000)]
atheros: v3.18: rename ATHEROS_AR231X symbol to ATH25

This patch starts upsteam ath25 code backporting.

No functional changes.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44706

9 years agoatheros: add v3.18 support
Felix Fietkau [Fri, 13 Mar 2015 02:57:41 +0000 (02:57 +0000)]
atheros: add v3.18 support

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44705

9 years agoglibc: make it more obvious that eglibc is a version of glibc
John Crispin [Thu, 12 Mar 2015 20:29:58 +0000 (20:29 +0000)]
glibc: make it more obvious that eglibc is a version of glibc

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44704

9 years agoglibc: remove accidentally commited folder
John Crispin [Thu, 12 Mar 2015 20:29:50 +0000 (20:29 +0000)]
glibc: remove accidentally commited folder

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44703

9 years agotoolchain: The glorious return of glibc, ver 2.21
John Crispin [Thu, 12 Mar 2015 19:50:57 +0000 (19:50 +0000)]
toolchain: The glorious return of glibc, ver 2.21

It's the eglibc packaging with a bit of spit-polishing. And testing. :-)

[blogic: merged glibc and eglibc into 1 and made eglibc a glibc variant]

Signed-off-by: Jeff Waugh <jdub@bethesignal.org>
Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44701

9 years agobusybox: Minor fix for glibc 2.21 compatibility
John Crispin [Thu, 12 Mar 2015 19:50:23 +0000 (19:50 +0000)]
busybox: Minor fix for glibc 2.21 compatibility

Signed-off-by: Jeff Waugh <jdub@bethesignal.org>
SVN-Revision: 44700

9 years agofstools: bump to latest git HEAD
John Crispin [Thu, 12 Mar 2015 19:50:14 +0000 (19:50 +0000)]
fstools: bump to latest git HEAD

this makes fstools compile with latest glibc

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44699

9 years agoprocd: bump to latest git HEAD
John Crispin [Thu, 12 Mar 2015 19:50:00 +0000 (19:50 +0000)]
procd: bump to latest git HEAD

this makes procd compile with latest glibc

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44698

9 years agoath9k: fix a beacon enable handling bug
Felix Fietkau [Thu, 12 Mar 2015 16:21:53 +0000 (16:21 +0000)]
ath9k: fix a beacon enable handling bug

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 44696

9 years agokernel: backport fib_trie improvements/fixes from 4.0-rc
Felix Fietkau [Thu, 12 Mar 2015 15:27:44 +0000 (15:27 +0000)]
kernel: backport fib_trie improvements/fixes from 4.0-rc

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 44695

9 years agouClibc: fix syscall() for syscalls with 6 arguments on PowerPC
John Crispin [Thu, 12 Mar 2015 10:07:01 +0000 (10:07 +0000)]
uClibc: fix syscall() for syscalls with 6 arguments on PowerPC

Patch from the uClibc master.

Fixes the package fastd on PowerPC.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
SVN-Revision: 44694

9 years agoar71xx: Ext LNA patch for TUBE2H and HORNET-UB targets
John Crispin [Thu, 12 Mar 2015 10:06:53 +0000 (10:06 +0000)]
ar71xx: Ext LNA patch for TUBE2H and HORNET-UB targets

Signed-off-by: Christophe Prévotaux <c.prevotaux@rural-networks.com>
SVN-Revision: 44693

9 years agoar71xx: Hornet UB GPIO WPS/Reset
John Crispin [Thu, 12 Mar 2015 10:06:42 +0000 (10:06 +0000)]
ar71xx: Hornet UB GPIO WPS/Reset

This problem has existed at least since Attitude Adjustment and
is also present in trunk. Basically on the Hornet-UB board the
functionality of RESET and WPS have "switched places".

There are two tickets about the issue at dev.openwrt.org,
The solution suggested on them both is incomplete though
and introduces the following proglem:

Patching as suggested on #14136/#15282 will result in a situation
where simply pressing the RESET button on the bottom will cause
FACTORY RESET to be run. This is due to GPIO high/low state being
incorrect as a result of the above change and virtually the RESET
button is in the pressed-down state the entire time. When it is
then physically pressed, that causes the opposite, release, to be
triggered and since to the board it seemed that the button was
pressed long before it was released, the FACTORY RESET results.

The attached patch works as expected. I have verified both the
incorrect functionality as well as after fixing the issue as
described in the patch and flashing the resulting firmware to a
Hornet-UB board.

Signed-off-by: Janne Cederberg <janne.cederberg@gmail.com>
SVN-Revision: 44692

9 years agoralink: resize the flash partition for FireWRT
John Crispin [Thu, 12 Mar 2015 10:06:31 +0000 (10:06 +0000)]
ralink: resize the flash partition for FireWRT

Signed-off-by: wengbj <fl.service@t-firefly.com>
SVN-Revision: 44691

9 years agoralink: MT7621 add i2c controller driver
John Crispin [Thu, 12 Mar 2015 10:06:17 +0000 (10:06 +0000)]
ralink: MT7621 add i2c controller driver

ralink i2c driver is not working on MT7621 platform. Porting a new drivers from MTK's source code.

Signed-off-by: daixj <fl.service@t-firefly.com>
SVN-Revision: 44690

9 years agolldpd: add support for 'readonly_mode'
John Crispin [Thu, 12 Mar 2015 10:06:08 +0000 (10:06 +0000)]
lldpd: add support for 'readonly_mode'

Signed-off-by: Alexandru Ardelean <ardeleanalex@gmail.com>
SVN-Revision: 44689

9 years agomodules: set dwc2 gadget dependency for at91
John Crispin [Thu, 12 Mar 2015 10:06:01 +0000 (10:06 +0000)]
modules: set dwc2 gadget dependency for at91

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44688

9 years agoacx-mac80211: fails to build on cobalt
John Crispin [Thu, 12 Mar 2015 10:05:54 +0000 (10:05 +0000)]
acx-mac80211: fails to build on cobalt

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44687

9 years agocobalt: framebuffer device has moved
John Crispin [Thu, 12 Mar 2015 10:05:48 +0000 (10:05 +0000)]
cobalt: framebuffer device has moved

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44686

9 years agodisable silent-rules by default
Mirko Vogt [Thu, 12 Mar 2015 00:39:50 +0000 (00:39 +0000)]
disable silent-rules by default

OpenWrt hides verbose output by default,
regardless of automake silent-rules being en-/disabled.

If we enable verbose output for package builds (V=s)
however, we'd like to see as most as possible.

Signed-off-by: Mirko Vogt <mirko@openwrt.org>
SVN-Revision: 44685

9 years agoDo not use $(V) - force AM_V=1
Mirko Vogt [Thu, 12 Mar 2015 00:39:48 +0000 (00:39 +0000)]
Do not use $(V) - force AM_V=1

Projects using silent-rules might otherwise fail to compile.

This is due to the following resulting Makefile code:

AM_V_P = $(am__v_P_$(V))
am__v_P_ = $(am__v_P_$(AM_DEFAULT_VERBOSITY))
am__v_P_0 = false
am__v_P_1 = :

Automake directly uses $(V) for variable name expansion
which fails if $(V) is set to sth. else other than '0' or '1'.

This patch forces automake to not take $(V) into account
but to always use the default config for verbosity.

A better approach would be to actually take $(V) into account,
however not just pass it through (AM_V='$(V)') but
set AM_V to 0 if $V=0 / unset and to 1 otherwise.

This plan however is foiled due to my frustration about
automake as well as my lack of m4 skills.

automake bug report: http://debbugs.gnu.org/cgi/bugreport.cgi?bug=20077

Signed-off-by: Mirko Vogt <mirko@openwrt.org>
SVN-Revision: 44684

9 years agomodules: make dwc2 only depend on gadget support for brcm2708
John Crispin [Wed, 11 Mar 2015 17:58:53 +0000 (17:58 +0000)]
modules: make dwc2 only depend on gadget support for brcm2708

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44683

9 years agouhttp: update to latest git HEAD
John Crispin [Wed, 11 Mar 2015 17:58:47 +0000 (17:58 +0000)]
uhttp: update to latest git HEAD

this add json-c 0.12, sorry forgot to push this earlier today

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44682

9 years agokernel: backport symbol export from r44653 to 3.14
Felix Fietkau [Wed, 11 Mar 2015 17:32:29 +0000 (17:32 +0000)]
kernel: backport symbol export from r44653 to 3.14

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 44681

9 years agobcm53xx: fixup early device id 8012
Rafał Miłecki [Wed, 11 Mar 2015 17:32:02 +0000 (17:32 +0000)]
bcm53xx: fixup early device id 8012

Looks like the BCM53012 has a similar problem to the BCM53011.

Signed-off-by: Ian Kent <raven@themaw.net>
SVN-Revision: 44680

9 years agobcm53xx: fix typo in bcm47xx sprom driver
Rafał Miłecki [Wed, 11 Mar 2015 17:10:19 +0000 (17:10 +0000)]
bcm53xx: fix typo in bcm47xx sprom driver

Fix thinko' in the bcm47xx sprom driver.

Signed-off-by: Ian Kent <raven@themaw.net>
SVN-Revision: 44679

9 years agokenrel: refresh patches
John Crispin [Wed, 11 Mar 2015 17:08:46 +0000 (17:08 +0000)]
kenrel: refresh patches

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44678

9 years agolantiq: make the new dwc2 support only work for vr9
John Crispin [Wed, 11 Mar 2015 17:08:40 +0000 (17:08 +0000)]
lantiq: make the new dwc2 support only work for vr9

danube needs to be added

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44677

9 years agolantiq: Convert Zyxel P-2812HNU-FX and TP-Link TD-W8970 to support dwc2
John Crispin [Wed, 11 Mar 2015 17:08:32 +0000 (17:08 +0000)]
lantiq: Convert Zyxel P-2812HNU-FX and TP-Link TD-W8970 to support dwc2

Here the device tree entry for ifxhcd is listed as compatible with one
supported in dwc2 (after patching the dwc driver appropriately).

A second entry is added to support the second core of the hcd. This
entry is listed to be compatible with only dwc2. Done this way there
should be backwards support for both hcd drivers (ltq-hcd and dwc2)

Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
Signed-off-by: Vincent Pelletier <plr.vincent@gmail.com>
SVN-Revision: 44676

9 years agolantiq: Use platform endianness when accessing dwc2 registers
John Crispin [Wed, 11 Mar 2015 17:08:26 +0000 (17:08 +0000)]
lantiq: Use platform endianness when accessing dwc2 registers

This patch switches calls to readl/writel to their
dwc2_readl/dwc2_writel equivalents which preserve platform endianness.

This patch is necessary to access dwc2 registers correctly on big
endian systems such as the mips based SoCs made by Lantiq. Then dwc2
can be used to replace ifx-hcd driver for Lantiq platforms found e.g.
in OpenWrt.

The patch was autogenerated with the following commands:
$EDITOR core.h
sed -i "s/\<readl\>/dwc2_readl/g" *.c hcd.h hw.h
sed -i "s/\<writel\>/dwc2_writel/g" *.c hcd.h hw.h

Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
Signed-off-by: Vincent Pelletier <plr.vincent@gmail.com>
SVN-Revision: 44675

9 years agolantiq: Add sensible hw-defaults for dwc2
John Crispin [Wed, 11 Mar 2015 17:08:15 +0000 (17:08 +0000)]
lantiq: Add sensible hw-defaults for dwc2

Lantiq driver does not work with autodetected fifo sizes so use ones
from original ltq-hcd driver in dwc2. Other values can be
autodetected.

Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
Signed-off-by: Vincent Pelletier <plr.vincent@gmail.com>
SVN-Revision: 44674

9 years agolantiq: Configure gpio power output pin when initializing dwc2 usb
John Crispin [Wed, 11 Mar 2015 17:08:08 +0000 (17:08 +0000)]
lantiq: Configure gpio power output pin when initializing dwc2 usb

Port gpio code from original ltq-hcd driver to dwc2.

Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
Signed-off-by: Vincent Pelletier <plr.vincent@gmail.com>
SVN-Revision: 44673

9 years agolantiq: Add usb initialization bits from ltq-hcd to platform init
John Crispin [Wed, 11 Mar 2015 17:08:02 +0000 (17:08 +0000)]
lantiq: Add usb initialization bits from ltq-hcd to platform init

Add VR9 specific usb initialization bits from ltq-hcd to platform
initialization.

This patch is more of a proof-of-concept than production quality
since the initialization registers are different on other lantiq
platforms.

Signed-off-by: Antti Seppälä <a.seppala@gmail.com>
Signed-off-by: Vincent Pelletier <plr.vincent@gmail.com>
SVN-Revision: 44672

9 years agoipset: Bump to 6.24
John Crispin [Wed, 11 Mar 2015 16:41:58 +0000 (16:41 +0000)]
ipset: Bump to 6.24

Included you'll find a patch to bump ipset to version 6.24. This
version supports the ip,fwmark set, which is needed for mwan3 1.6.

Signed-off-by: Jeroen Louwes <jeroen.louwes@gmail.com>
SVN-Revision: 44671

9 years agomac80211: add missing header file
Felix Fietkau [Wed, 11 Mar 2015 16:36:35 +0000 (16:36 +0000)]
mac80211: add missing header file

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 44670

9 years agomac80211: remove obsolete patches
Felix Fietkau [Wed, 11 Mar 2015 16:25:09 +0000 (16:25 +0000)]
mac80211: remove obsolete patches

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 44669

9 years agocobalt: rename config file
John Crispin [Wed, 11 Mar 2015 16:22:01 +0000 (16:22 +0000)]
cobalt: rename config file

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44668

9 years agorpcd: update to latest git HEAD
John Crispin [Wed, 11 Mar 2015 15:56:51 +0000 (15:56 +0000)]
rpcd: update to latest git HEAD

adds 3 fixes from luka

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44666

9 years agoar71xx: improve WD's My Net Wi-fi Range Extender image creation
John Crispin [Wed, 11 Mar 2015 15:56:34 +0000 (15:56 +0000)]
ar71xx: improve WD's My Net Wi-fi Range Extender image creation

Previously, the generated images for the My Net Wi-fi Range Extender
wouldn't always work (and panic) due to the fixed mtd offsets and
sizes for the kernel and rootfs. This patch fixes the problem by
utilizing the shared Cybertan's partition parser to recalculate
the mtd partitions for every image dynamically everytime.

Reported-by: Pascal Paradis <peparadis@yahoo.com>
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
SVN-Revision: 44665

9 years agoar71xx: generalize cybertan partition parser
John Crispin [Wed, 11 Mar 2015 15:56:27 +0000 (15:56 +0000)]
ar71xx: generalize cybertan partition parser

By removing the NL16 signature check, the parser can be
utilized by other devices like the WD My Net Wi-fi Range
Extender.

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
SVN-Revision: 44664

9 years agoar71xx: rename wrt160nl's trx partition parser
John Crispin [Wed, 11 Mar 2015 15:56:15 +0000 (15:56 +0000)]
ar71xx: rename wrt160nl's trx partition parser

This patch renames the partition parser from
wrt160nl to more generic cybertan.

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
SVN-Revision: 44663

9 years agoar71xx: fix WRT160NL and TP-Link partition parser patches
John Crispin [Wed, 11 Mar 2015 15:55:55 +0000 (15:55 +0000)]
ar71xx: fix WRT160NL and TP-Link partition parser patches

405-mtd-tp-link-partition-parser.patch was updating the
WRT160NL's partition parser Kconfig. This patch moves the
relevant change into the right file:
404-mtd-wrt160nl-trx-parser.patch

Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
SVN-Revision: 44662

9 years agoralink: FireWRT flash size is up to 16M.
John Crispin [Wed, 11 Mar 2015 15:55:28 +0000 (15:55 +0000)]
ralink: FireWRT flash size is up to 16M.

Signed-off-by: wengbj <fl.service@t-firefly.com>
SVN-Revision: 44661

9 years agoralink: add FireWRT IO multiplexing
John Crispin [Wed, 11 Mar 2015 15:55:16 +0000 (15:55 +0000)]
ralink: add FireWRT IO multiplexing

Signed-off-by: wengbj <fl.service@t-firefly.com>
SVN-Revision: 44660

9 years agoar71xx: fix switched WLAN LEDs on TP-LINK Archer C5/C7
John Crispin [Wed, 11 Mar 2015 15:55:06 +0000 (15:55 +0000)]
ar71xx: fix switched WLAN LEDs on TP-LINK Archer C5/C7

ath10k is loaded before ath9k, so the 5GHz adapter becomes phy0.

Signed-off-by: Matthias Schiffer <mschiffer@universe-factory.net>
SVN-Revision: 44659

9 years agoFix 3.18.8 breakage of UBI devices with EOF marker (e.g. WNDR4300)
John Crispin [Wed, 11 Mar 2015 15:54:52 +0000 (15:54 +0000)]
Fix 3.18.8 breakage of UBI devices with EOF marker (e.g. WNDR4300)

This commit re-adds a patch from 3.14 that is required for UBI block
devices with an EOF marker to be successfully mounted.

Signed-off-by: Oliver Smith <oliver@8.c.9.b.0.7.4.0.1.0.0.2.ip6.arpa>
SVN-Revision: 44658

9 years agojson-c: update to 0.12 and bump all depending services
John Crispin [Wed, 11 Mar 2015 15:54:33 +0000 (15:54 +0000)]
json-c: update to 0.12 and bump all depending services

Version 0.12 deprecates json_object_object_get and moves the header files around

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44657

9 years agomac80211: Use newer firmware for ath10k
Felix Fietkau [Wed, 11 Mar 2015 15:02:54 +0000 (15:02 +0000)]
mac80211: Use newer firmware for ath10k

Patch by Bryan Forbes <bryan@reigndropsfall.net>

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 44656

9 years agomac80211: update to wireless-testing 2015-03-09
Felix Fietkau [Wed, 11 Mar 2015 15:02:47 +0000 (15:02 +0000)]
mac80211: update to wireless-testing 2015-03-09

Based on patch by Bryan Forbes <bryan@reigndropsfall.net>

Also update mt76 to update for API changes

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 44655

9 years agoiw: update to 3.17
Felix Fietkau [Wed, 11 Mar 2015 15:02:32 +0000 (15:02 +0000)]
iw: update to 3.17

Patch by Bryan Forbes <bryan@reigndropsfall.net>

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 44654

9 years agokernel: backport an export required by newer compat-wireless versions
Felix Fietkau [Wed, 11 Mar 2015 15:02:16 +0000 (15:02 +0000)]
kernel: backport an export required by newer compat-wireless versions

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 44653

9 years agoppc44x: use the mtdsplit framework for the Taishan
Imre Kaloz [Wed, 11 Mar 2015 12:20:03 +0000 (12:20 +0000)]
ppc44x: use the mtdsplit framework for the Taishan

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 44652

9 years agobase-files: firstboot: pass down arguments such as -y to jffs2reset
Luka Perkov [Tue, 10 Mar 2015 23:43:25 +0000 (23:43 +0000)]
base-files: firstboot: pass down arguments such as -y to jffs2reset

This brings back old behavior.

Signed-off-by: Luka Perkov <luka@openwrt.org>
SVN-Revision: 44651

9 years agoar71xx: fix gpio-latch driver
John Crispin [Tue, 10 Mar 2015 20:19:54 +0000 (20:19 +0000)]
ar71xx: fix gpio-latch driver

gpio api has changed. the remove call no longer returns a value.

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44650

9 years agoar71xx: refresh patches
Luka Perkov [Tue, 10 Mar 2015 16:11:44 +0000 (16:11 +0000)]
ar71xx: refresh patches

Signed-off-by: Luka Perkov <luka@openwrt.org>
SVN-Revision: 44649

9 years agoar71xx: wpj588: add missing usb support
Luka Perkov [Tue, 10 Mar 2015 16:11:39 +0000 (16:11 +0000)]
ar71xx: wpj588: add missing usb support

Signed-off-by: Luka Perkov <luka@openwrt.org>
Tested-By: Christian Mehlis <christian@m3hlis.de>
SVN-Revision: 44648

9 years agoar71xx: Fix LED polarity for the TP-LINK TL-MR13U.
John Crispin [Tue, 10 Mar 2015 13:26:50 +0000 (13:26 +0000)]
ar71xx: Fix LED polarity for the TP-LINK TL-MR13U.

Please also backport to Barrier Breaker (this same patch applies there too).

Signed-off-by: Vittorio Gambaletta <openwrt@vittgam.net>
SVN-Revision: 44647

9 years agoar71xx: Fix board detection for the TP-LINK TL-MR13U.
John Crispin [Tue, 10 Mar 2015 13:26:44 +0000 (13:26 +0000)]
ar71xx: Fix board detection for the TP-LINK TL-MR13U.

Signed-off-by: Vittorio Gambaletta <openwrt@vittgam.net>
SVN-Revision: 44646

9 years agokernel: fix yaffs compile for >= v3.16
John Crispin [Tue, 10 Mar 2015 13:08:17 +0000 (13:08 +0000)]
kernel: fix yaffs compile for >= v3.16

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44643

9 years agoralink: add missing symbol that broke full builds
John Crispin [Tue, 10 Mar 2015 11:03:16 +0000 (11:03 +0000)]
ralink: add missing symbol that broke full builds

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44642

9 years agokirkwood: add missing symbol that broke full builds
John Crispin [Tue, 10 Mar 2015 11:03:10 +0000 (11:03 +0000)]
kirkwood: add missing symbol that broke full builds

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44641

9 years agoau1000: build fails if the input layer is not selected
John Crispin [Tue, 10 Mar 2015 11:03:04 +0000 (11:03 +0000)]
au1000: build fails if the input layer is not selected

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44640

9 years agoipq806x: add 2 missing symbols that broke a full build
John Crispin [Tue, 10 Mar 2015 11:02:52 +0000 (11:02 +0000)]
ipq806x: add 2 missing symbols that broke a full build

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44639

9 years agoar7: register_mtd_parser() does no longer have a return value
John Crispin [Tue, 10 Mar 2015 11:02:45 +0000 (11:02 +0000)]
ar7: register_mtd_parser() does no longer have a return value

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 44638

9 years agoppc44x: upgrade and switch to 3.18
Imre Kaloz [Tue, 10 Mar 2015 10:00:41 +0000 (10:00 +0000)]
ppc44x: upgrade and switch to 3.18

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 44637

9 years ago6rd: make host dependency more specific
Steven Barth [Tue, 10 Mar 2015 08:43:54 +0000 (08:43 +0000)]
6rd: make host dependency more specific

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 44636

9 years ago6in4: make host dependency more specific
Steven Barth [Tue, 10 Mar 2015 08:43:40 +0000 (08:43 +0000)]
6in4: make host dependency more specific

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 44635

9 years agoebtables: revert "disable rpath" (closes: #19088)
Nicolas Thill [Mon, 9 Mar 2015 22:07:05 +0000 (22:07 +0000)]
ebtables: revert "disable rpath" (closes: #19088)

This reverts commit 9340723d70bce9d905c3f53b2bf415963153b34d.

Signed-off-by: Nicolas Thill <nico@openwrt.org>
SVN-Revision: 44634

9 years agokmod-usb-dwc2: depends on udc-core (kmod-usb-gadget)
John Crispin [Mon, 9 Mar 2015 14:07:30 +0000 (14:07 +0000)]
kmod-usb-dwc2: depends on udc-core (kmod-usb-gadget)

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 44633

9 years agolantiq: ltq-ptm-{ar9, ase, xway}: fix alloc_netdev for 3.18
John Crispin [Mon, 9 Mar 2015 14:07:14 +0000 (14:07 +0000)]
lantiq: ltq-ptm-{ar9, ase, xway}: fix alloc_netdev for 3.18

buildbot complained about
ifxmips_ptm_adsl.c:1445:70: error: macro "alloc_netdev" requires 4 arguments, but only 3 given
         g_net_dev[i] = alloc_netdev(0, g_net_dev_name[i], ether_setup);
                                                                      ^
fix that similar to the change made to ltq_ptm_vdsl.c
compile tested only.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 44632

9 years agoncm, qmi, mbim: Add dependency on wwan package
John Crispin [Mon, 9 Mar 2015 14:07:06 +0000 (14:07 +0000)]
ncm, qmi, mbim: Add dependency on wwan package

The wwan package holds the hotplug script to set mobile broadband
interfaces (un)available. Add it as a dependency to comgt-ncm,
uqmi and umbim.

Signed-off-by: Matti Laakso <malaakso@elisanet.fi>
SVN-Revision: 44631

9 years agoncm: Remove unnecessary proto_set_available commands
John Crispin [Mon, 9 Mar 2015 14:06:50 +0000 (14:06 +0000)]
ncm: Remove unnecessary proto_set_available commands

Interface should not be set unavailable in all error cases,
returning 1 is enough.

Signed-off-by: Matti Laakso <malaakso@elisanet.fi>
SVN-Revision: 44630

9 years agowwan: Improve USB modem hotplug script
John Crispin [Mon, 9 Mar 2015 14:06:30 +0000 (14:06 +0000)]
wwan: Improve USB modem hotplug script

Adds ncm proto to the list of checked protocols, sets interfaces
unavailable on device removal and removes the unnecessary ifup
command.

Signed-off-by: Matti Laakso <malaakso@elisanet.fi>
SVN-Revision: 44629

9 years agocomgt, wwan: Move USB data files to wwan package
John Crispin [Mon, 9 Mar 2015 14:06:20 +0000 (14:06 +0000)]
comgt, wwan: Move USB data files to wwan package

USB modem data files should be a part of the wwan package, which
actually uses them unlike comgt.

Signed-off-by: Matti Laakso <malaakso@elisanet.fi>
SVN-Revision: 44628

9 years agoutil-linux: fix pkgconfig files
Nicolas Thill [Mon, 9 Mar 2015 02:02:50 +0000 (02:02 +0000)]
util-linux: fix pkgconfig files

Upstream choose to not rely on configure.ac and recursively expand paths in pkgconfig files at make time,
leading to absolute /usr/include & /usr/lib in there... And we don't want that when cross-compiling...

Signed-off-by: Nicolas Thill <nico@openwrt.org>
SVN-Revision: 44627

9 years agobcm53xx: fix printing NAND controller revision
Rafał Miłecki [Sat, 7 Mar 2015 12:31:42 +0000 (12:31 +0000)]
bcm53xx: fix printing NAND controller revision

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
SVN-Revision: 44622

9 years agomvebu: add backported patch to reenable hw i/o coherency
Imre Kaloz [Fri, 6 Mar 2015 18:57:49 +0000 (18:57 +0000)]
mvebu: add backported patch to reenable hw i/o coherency

Reported-by: Claudio Leite <leitec@staticky.com>
Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 44621

9 years agoar71xx: add support for compex wpj558
Luka Perkov [Fri, 6 Mar 2015 14:41:57 +0000 (14:41 +0000)]
ar71xx: add support for compex wpj558

OpenWrt can be flashed with following uboot commands:

tftpboot 0x80500000 openwrt-ar71xx-generic-wpj558-16M-squashfs-sysupgrade.bin
erase 0x9f030000 +$filesize
cp.b $fileaddr 0x9f030000 $filesize

Signed-off-by: Luka Perkov <luka@openwrt.org>
SVN-Revision: 44620

9 years agoodhcpd: improvements for DHCPv4 and compile fixes
Steven Barth [Fri, 6 Mar 2015 14:41:07 +0000 (14:41 +0000)]
odhcpd: improvements for DHCPv4 and compile fixes

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 44619

9 years agoopenssl: enable ARM assembly acceleration
John Crispin [Fri, 6 Mar 2015 07:57:10 +0000 (07:57 +0000)]
openssl: enable ARM assembly acceleration

Tested myself on ixp4xx and mvebu, and (originally)
by Daniel on i.MX6. Also tested on a MIPS target,
to make sure the change to ASFLAGS does not break things.

Based on a patch submitted by Daniel Drown:

https://lists.openwrt.org/pipermail/openwrt-devel/2014-July/026639.html

Signed-off-by: Claudio Leite <leitec@staticky.com>
Signed-off-by: Daniel Drown <dan-openwrt@drown.org>
SVN-Revision: 44618

9 years agoswconfig: fix build with linux 4.0
John Crispin [Fri, 6 Mar 2015 07:57:03 +0000 (07:57 +0000)]
swconfig: fix build with linux 4.0

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 44617

9 years agoipq806x: rename patches in patches-3.18
John Crispin [Fri, 6 Mar 2015 07:56:58 +0000 (07:56 +0000)]
ipq806x: rename patches in patches-3.18

This will allow ipq806x to support multiple kernel version more easily.

Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 44616

9 years agokernel: move KERNFS to generic config
John Crispin [Fri, 6 Mar 2015 07:56:47 +0000 (07:56 +0000)]
kernel: move KERNFS to generic config

KERNFS symbol is selected by SYSFS, so place it in generic config.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44615

9 years agokernel: disable ARCH_NEEDS_CPU_IDLE_COUPLED
John Crispin [Fri, 6 Mar 2015 07:56:34 +0000 (07:56 +0000)]
kernel: disable ARCH_NEEDS_CPU_IDLE_COUPLED

Disable ARCH_NEEDS_CPU_IDLE_COUPLED by-default in generic config, since
only one platfrom (omap) needs them.

Signed-off-by: Sergey Ryazanov <ryazanov.s.a@gmail.com>
SVN-Revision: 44614