openwrt/svn-archive/archive.git
6 years ago[rules.mk] remove duplicate HOSTCC_NOCACHE
John Crispin [Fri, 10 Aug 2012 19:07:03 +0000 (19:07 +0000)]
[rules.mk] remove duplicate HOSTCC_NOCACHE

SVN-Revision: 33088

6 years ago[tools] fix ccache build
John Crispin [Fri, 10 Aug 2012 19:05:22 +0000 (19:05 +0000)]
[tools] fix ccache build

SVN-Revision: 33087

6 years ago[buildroot] move LIB_SUFFIX from base-files to rules.mk
Florian Fainelli [Fri, 10 Aug 2012 14:15:06 +0000 (14:15 +0000)]
[buildroot] move LIB_SUFFIX from base-files to rules.mk

Some packages are going to need fixups based on LIB_SUFFIX

SVN-Revision: 33080

6 years ago[mcs814x] nuport-mac: delay call to napi_start() until fully able to process RX
Florian Fainelli [Fri, 10 Aug 2012 07:04:21 +0000 (07:04 +0000)]
[mcs814x] nuport-mac: delay call to napi_start() until fully able to process RX

SVN-Revision: 33079

6 years ago[mcs814x] set a compatible command-line to boot from flash
Florian Fainelli [Fri, 10 Aug 2012 07:04:16 +0000 (07:04 +0000)]
[mcs814x] set a compatible command-line to boot from flash

SVN-Revision: 33078

6 years agogeneric: rtl8366_smi: increase delay after switch hw reset
Gabor Juhos [Fri, 10 Aug 2012 06:32:00 +0000 (06:32 +0000)]
generic: rtl8366_smi: increase delay after switch hw reset

SVN-Revision: 33077

6 years ago[package] netifd: call "wifi down" before "wifi up"
Jo-Philipp Wich [Thu, 9 Aug 2012 22:44:52 +0000 (22:44 +0000)]
[package] netifd: call "wifi down" before "wifi up"

When the underlying /etc/config/wireless got changed since the last "wifi up"
has been performed, the uci vap ifname state vars become inconsistent on a
subsequent "wifi up" and multiple vaps get mapped to the same ifnames which
confuses the gui and other processes relying on them.

For now call an explicit "wifi down" prior to each "wifi up" which will clear
up the state accordingly.

SVN-Revision: 33076

6 years ago[package] base-files: fix option parsing of -F/--force parameter
Jo-Philipp Wich [Thu, 9 Aug 2012 18:30:47 +0000 (18:30 +0000)]
[package] base-files: fix option parsing of -F/--force parameter

SVN-Revision: 33075

6 years agopackage: linux-atm: br2684ct: fix nas interface bringup on boot
Jonas Gorski [Thu, 9 Aug 2012 13:13:26 +0000 (13:13 +0000)]
package: linux-atm: br2684ct: fix nas interface bringup on boot

Give nas interface time to setup them selves before configuring them.
Should fix #11826.

Contributed by T-Labs, Deutsche Telekom Innovation Laboratories

Signed-off-by: Jonas Gorski <jonas.gorski@gmail.com>
SVN-Revision: 33074

6 years agoFix hard reset of alix/soekris hardware under heavy NIC load
Jo-Philipp Wich [Thu, 9 Aug 2012 09:41:22 +0000 (09:41 +0000)]
Fix hard reset of alix/soekris hardware under heavy NIC load

The following patch corrects a hard reset that occurs on alix and
soekris net5501 hardware. Specifically, this is an issue with the via
rhine NIC driver. Under periods of extreme load, the via rhine driver
can cause a reset of the entire system. When this happens no output is
seen on the console, the device simply reboots. This was reported in the
following ticket:

Consistent crash on Soekris NET5501
https://dev.openwrt.org/ticket/11882

I was able to reproduce the problem locally with numerous alix 2d13
platforms. Under heavy 64 byte frame size load from a Smartbits traffic
generator, the alix hardware will hard reset in under 5 minutes. The
same hardware, under the same load, on FreeBSD 9.0 does not reset.

The patch below backports most of the via rhine changes from kernel 3.5.
With OpenWrt trunk being on kernel 3.3.8, it seemed prudent to bring
these changes in to avoid this serious issue.

I've tested this heavily in my test environment and was able to keep the
device stable under load for extended periods of time without any resets.

Also, a nice side effect of this change is that it significantly raises
the performance the platform. Without this patch the alix 2d13 can move
approximately 29,000 packets per second at 64 byte frame sizes. After
this patch the alix can move 52,000 packets per second at 64 byte frame
sizes.

I put this patch under the x86 patches as those are the only platforms
with via rhine hardware that I'm aware of. If it needs to go some place
else please let me know.

Thanks,
Adam

Signed-off-by: Adam Gensler <openwrt@kristenandadam.net>
SVN-Revision: 33072

6 years ago[mcs814x] nuport-mac: use dma_mapping_error() instead of checking pointers
Florian Fainelli [Thu, 9 Aug 2012 09:06:35 +0000 (09:06 +0000)]
[mcs814x] nuport-mac: use dma_mapping_error() instead of checking pointers

Thanks Gabor for spotting this

SVN-Revision: 33071

6 years ago[mcs814x] nuport-mac: rework PHY interrupt configuration to avoid races
Florian Fainelli [Wed, 8 Aug 2012 21:51:18 +0000 (21:51 +0000)]
[mcs814x] nuport-mac: rework PHY interrupt configuration to avoid races

SVN-Revision: 33070

6 years ago[mcs814x] nuport-mac: close possible race condition during start
Florian Fainelli [Wed, 8 Aug 2012 21:51:13 +0000 (21:51 +0000)]
[mcs814x] nuport-mac: close possible race condition during start

SVN-Revision: 33069

6 years ago[mcs814x] nuport-mac: unmap the DMA buffer when closing device
Florian Fainelli [Wed, 8 Aug 2012 21:51:09 +0000 (21:51 +0000)]
[mcs814x] nuport-mac: unmap the DMA buffer when closing device

SVN-Revision: 33068

6 years ago[mcs814x] nuport-mac: check dma_map_single address return values
Florian Fainelli [Wed, 8 Aug 2012 21:51:04 +0000 (21:51 +0000)]
[mcs814x] nuport-mac: check dma_map_single address return values

SVN-Revision: 33067

6 years ago[mcs814x] build a Devolo dLAN USB Extender specific image
Florian Fainelli [Wed, 8 Aug 2012 21:51:00 +0000 (21:51 +0000)]
[mcs814x] build a Devolo dLAN USB Extender specific image

The kernel has to be rounded up to the nearest 1MB, then we append
the root filesystem at its end.

SVN-Revision: 33066

6 years ago[mcs814x] cleanup image generation
Florian Fainelli [Wed, 8 Aug 2012 21:50:55 +0000 (21:50 +0000)]
[mcs814x] cleanup image generation

SVN-Revision: 33065

6 years agoath9k: fix handling of decryption errors, should improve link stability
Felix Fietkau [Wed, 8 Aug 2012 18:23:04 +0000 (18:23 +0000)]
ath9k: fix handling of decryption errors, should improve link stability

SVN-Revision: 33063

6 years agoath5k: revert a buggy tx power related patch
Felix Fietkau [Wed, 8 Aug 2012 18:23:00 +0000 (18:23 +0000)]
ath5k: revert a buggy tx power related patch

SVN-Revision: 33062

6 years ago[package] base-files: check for ldd presence and executability (#11991)
Florian Fainelli [Wed, 8 Aug 2012 15:37:34 +0000 (15:37 +0000)]
[package] base-files: check for ldd presence and executability (#11991)

SVN-Revision: 33057

6 years agoath9k: fix hang issues on hw reset caused by interrupt storms
Felix Fietkau [Wed, 8 Aug 2012 12:56:44 +0000 (12:56 +0000)]
ath9k: fix hang issues on hw reset caused by interrupt storms

SVN-Revision: 33055

6 years ago[lantiq] pci gpio mapping
John Crispin [Wed, 8 Aug 2012 12:14:43 +0000 (12:14 +0000)]
[lantiq] pci gpio mapping

This patch fixes the GPIO ALTSEL settings for some of the GPIOs used by the PCI subsystem in Lantiq Danube. These changes are required for more than one PCI device to work. Tested with an ARV7510PW having two PCI-devices; a VIA USB controller and a Ralink WLAN mini-PCI card.

Signed-off-by: Matti Laakso <malaakso@elisanet.fi>
SVN-Revision: 33054

6 years ago[package] hostapd: fix ap/sta regression which got accidentally introduced by r32829...
Jo-Philipp Wich [Wed, 8 Aug 2012 11:13:12 +0000 (11:13 +0000)]
[package] hostapd: fix ap/sta regression which got accidentally introduced by r32829 (#11914)

SVN-Revision: 33053

6 years ago[package] busybox: support V=c toplevel verboseness flag (#11799)
Florian Fainelli [Tue, 7 Aug 2012 16:08:11 +0000 (16:08 +0000)]
[package] busybox: support V=c toplevel verboseness flag (#11799)

Signed-off-by: Gregory M. Turner <gmt@malth.us>
SVN-Revision: 33035

6 years ago[package] w1-gpio-custom: set enable_external_pullup to null (#8952)
Florian Fainelli [Tue, 7 Aug 2012 16:08:08 +0000 (16:08 +0000)]
[package] w1-gpio-custom: set enable_external_pullup to null (#8952)

SVN-Revision: 33034

6 years agoath9k: completely disable the MIB interrupt from the interrupt mask register to preve...
Felix Fietkau [Tue, 7 Aug 2012 09:36:24 +0000 (09:36 +0000)]
ath9k: completely disable the MIB interrupt from the interrupt mask register to prevent potential interrupt storms

SVN-Revision: 33032

6 years agoath9k: extend and enable PA predistortion
Felix Fietkau [Mon, 6 Aug 2012 22:11:30 +0000 (22:11 +0000)]
ath9k: extend and enable PA predistortion

SVN-Revision: 33031

6 years agomac80211: do not delete linux/*_platform.h, they are touched by patches. instead...
Felix Fietkau [Mon, 6 Aug 2012 20:18:07 +0000 (20:18 +0000)]
mac80211: do not delete linux/*_platform.h, they are touched by patches. instead make the build fail if they differ from the ones in the kernel tree to prevent subtle runtime issues (see #11981)

SVN-Revision: 33030

6 years ago[lantiq] fixes tapi dependencies
John Crispin [Mon, 6 Aug 2012 18:56:38 +0000 (18:56 +0000)]
[lantiq] fixes tapi dependencies

SVN-Revision: 33029

6 years ago[e2fsprogs] fixes package description
John Crispin [Mon, 6 Aug 2012 18:54:36 +0000 (18:54 +0000)]
[e2fsprogs] fixes package description

SVN-Revision: 33028

6 years agoath5k: add various pending tx power fixes, vastly improves stability and performance...
Felix Fietkau [Mon, 6 Aug 2012 15:33:24 +0000 (15:33 +0000)]
ath5k: add various pending tx power fixes, vastly improves stability and performance with various cards

SVN-Revision: 33014

6 years ago[packages] mac80211: remove header files used to share struct with platform code.
Hauke Mehrtens [Sun, 5 Aug 2012 23:43:36 +0000 (23:43 +0000)]
[packages] mac80211: remove header files used to share struct with platform code.

The structs defined in these header files should be the same as the
structes used by the platform code otherwise it will cause strange
runtime errors. By removing them from compat-wireless we will get a
compile error instead of a runtime error if they are out of sync.

SVN-Revision: 33006

6 years agobcm63xx: fix USB base registers and IRQs for BCM6328
Jonas Gorski [Sun, 5 Aug 2012 23:18:22 +0000 (23:18 +0000)]
bcm63xx: fix USB base registers and IRQs for BCM6328

SVN-Revision: 33005

6 years ago[brcm47xx] update tg3 driver
Hauke Mehrtens [Sun, 5 Aug 2012 20:27:40 +0000 (20:27 +0000)]
[brcm47xx] update tg3 driver

SVN-Revision: 33004

6 years ago[brcm47xx] generate image for Linksys e3200
Hauke Mehrtens [Sun, 5 Aug 2012 13:14:00 +0000 (13:14 +0000)]
[brcm47xx] generate image for Linksys e3200

SVN-Revision: 33003

6 years ago[packages] switch: clean up robo switch driver
Hauke Mehrtens [Sun, 5 Aug 2012 13:10:43 +0000 (13:10 +0000)]
[packages] switch: clean up robo switch driver

 * remove use_et, all supported drivers are using mii ioctls
 * remove robo.phy_addr, phy_addr was always set to ROBO_PHY_ADDR.
 * remove support for old kernel versions
 * do_ioctl is allways called with buff == NULL
 * use if_mii

SVN-Revision: 33002

6 years agobrcm47xx: update gpio patch
Hauke Mehrtens [Sun, 5 Aug 2012 12:40:05 +0000 (12:40 +0000)]
brcm47xx: update gpio patch

SVN-Revision: 33001

6 years agobrcm47xx: fix patch which moves the flash struct in wgt634u code
Hauke Mehrtens [Sun, 5 Aug 2012 12:39:15 +0000 (12:39 +0000)]
brcm47xx: fix patch which moves the flash struct in wgt634u code

SVN-Revision: 33000

6 years ago[kernel] fix xen pci module, partially fixes #11970
John Crispin [Sun, 5 Aug 2012 08:16:50 +0000 (08:16 +0000)]
[kernel] fix xen pci module, partially fixes #11970

SVN-Revision: 32996

6 years ago[lantiq] W502V/ARV4525 wireless mac bugfix
John Crispin [Sat, 4 Aug 2012 20:14:52 +0000 (20:14 +0000)]
[lantiq] W502V/ARV4525 wireless mac bugfix

The ARV4525 board init is passing the wrong offset to
arv_register_ath5k() resulting in a bogus mac.

Signed-off-by: Tobias Diedrich <ranma+openwrt@tdiedrich.de>
SVN-Revision: 32993

6 years agobrcm47xx: use libgpio instaed of implementing the gpio interface ourself.
Hauke Mehrtens [Sat, 4 Aug 2012 19:46:25 +0000 (19:46 +0000)]
brcm47xx: use libgpio instaed of implementing the gpio interface ourself.

SVN-Revision: 32992

6 years ago[kernel] gspca-sn9c20x.ko was not insmod'ed due to typo
John Crispin [Sat, 4 Aug 2012 19:17:49 +0000 (19:17 +0000)]
[kernel] gspca-sn9c20x.ko was not insmod'ed due to typo

fixes #11973

SVN-Revision: 32991

6 years ago[atheros] wrt55agv2-spidevs needs <linux/modules.h>
John Crispin [Fri, 3 Aug 2012 09:38:24 +0000 (09:38 +0000)]
[atheros] wrt55agv2-spidevs needs <linux/modules.h>

SVN-Revision: 32956

6 years ago[lantiq] cleanup patches
John Crispin [Fri, 3 Aug 2012 08:53:02 +0000 (08:53 +0000)]
[lantiq] cleanup patches

SVN-Revision: 32953

6 years ago[mcs814x] nuport-mac: properly protect the phy interrupt handler
Florian Fainelli [Thu, 2 Aug 2012 18:13:17 +0000 (18:13 +0000)]
[mcs814x] nuport-mac: properly protect the phy interrupt handler

SVN-Revision: 32951

6 years ago[bufferbloat] Codel: avoid a nul rec_inv_sqrt
John Crispin [Thu, 2 Aug 2012 13:39:50 +0000 (13:39 +0000)]
[bufferbloat] Codel: avoid a nul rec_inv_sqrt

One condition before codel_Newton_step() was not good if
we never left the dropping state for a flow. As a result
rec_inv_sqrt was 0, instead of the ~0 initial value.

codel control law was then set to a very aggressive mode, dropping
many packets before reaching 'target' and recovering from this problem.

Brought over from 3.5-stable

Signed-off-by: Dave Taht <dave.taht@bufferbloat.net>
SVN-Revision: 32950

6 years ago[lantiq] fixes i2c clock enable
John Crispin [Thu, 2 Aug 2012 13:29:49 +0000 (13:29 +0000)]
[lantiq] fixes i2c clock enable

SVN-Revision: 32949

6 years agoar71xx: simplify misc IRQ handling
Gabor Juhos [Thu, 2 Aug 2012 11:25:49 +0000 (11:25 +0000)]
ar71xx: simplify misc IRQ handling

SVN-Revision: 32948

6 years agoar71xx: fix AR934X_EHCI_SIZE
Gabor Juhos [Thu, 2 Aug 2012 11:25:46 +0000 (11:25 +0000)]
ar71xx: fix AR934X_EHCI_SIZE

SVN-Revision: 32947

6 years agoar71xx: image: stop the switch from the lzma-loader on the TL-WR1043ND
Gabor Juhos [Thu, 2 Aug 2012 09:54:41 +0000 (09:54 +0000)]
ar71xx: image: stop the switch from the lzma-loader on the TL-WR1043ND

SVN-Revision: 32946

6 years agoar71xx: image: allow to use board specific code in the lzma-loader
Gabor Juhos [Thu, 2 Aug 2012 09:54:33 +0000 (09:54 +0000)]
ar71xx: image: allow to use board specific code in the lzma-loader

SVN-Revision: 32945

6 years agoar71xx: init switch hw_reset on the TL-WR1043ND board
Gabor Juhos [Thu, 2 Aug 2012 09:54:28 +0000 (09:54 +0000)]
ar71xx: init switch hw_reset on the TL-WR1043ND board

SVN-Revision: 32944

6 years agogeneric: rtl836x: add hw_reset field to struct rtl8366_smi
Gabor Juhos [Thu, 2 Aug 2012 09:54:21 +0000 (09:54 +0000)]
generic: rtl836x: add hw_reset field to struct rtl8366_smi

It will be used to start/stop the switch if that is
supported by the given board.

SVN-Revision: 32943

6 years agogeneric: rtl8366_smi: use __rtl8366_smi_cleanup instead of gpio_free calls
Gabor Juhos [Thu, 2 Aug 2012 09:54:13 +0000 (09:54 +0000)]
generic: rtl8366_smi: use __rtl8366_smi_cleanup instead of gpio_free calls

SVN-Revision: 32942

6 years ago[mcs814x] add missing netmask to the network config
Florian Fainelli [Wed, 1 Aug 2012 22:09:13 +0000 (22:09 +0000)]
[mcs814x] add missing netmask to the network config

SVN-Revision: 32941

6 years ago[mcs814x] assume the link is up and full-duplex by default
Florian Fainelli [Wed, 1 Aug 2012 21:56:54 +0000 (21:56 +0000)]
[mcs814x] assume the link is up and full-duplex by default

SVN-Revision: 32940

6 years ago[mcs814x] add a default and suitable network config file
Florian Fainelli [Wed, 1 Aug 2012 21:56:49 +0000 (21:56 +0000)]
[mcs814x] add a default and suitable network config file

SVN-Revision: 32939

6 years agoar71xx: ew-dorin: added router firmware generation
Gabor Juhos [Wed, 1 Aug 2012 12:24:57 +0000 (12:24 +0000)]
ar71xx: ew-dorin: added router firmware generation

[juhosg: remove dead code]

Signed-off-by: embedded wireless GmbH <info@embeddedwireless.de>
SVN-Revision: 32934

6 years agoar71xx: ew-dorin: added USB support
Gabor Juhos [Wed, 1 Aug 2012 12:24:56 +0000 (12:24 +0000)]
ar71xx: ew-dorin: added USB support

Signed-off-by: embedded wireless GmbH <info@embeddedwireless.de>
SVN-Revision: 32933

6 years agoar71xx: Buffalo WLAE-AG300N wireless led patch
Gabor Juhos [Wed, 1 Aug 2012 12:24:54 +0000 (12:24 +0000)]
ar71xx: Buffalo WLAE-AG300N wireless led patch

Here is the uci-defaults version of the led support for
the Buffalo WLAE-AG300N.

Signed-off-by: Yoichi Shinoda <shinoda@jaist.ac.jp>
SVN-Revision: 32932

6 years agoar71xx: Buffalo WLAE-AG300N initial support
Gabor Juhos [Wed, 1 Aug 2012 12:24:52 +0000 (12:24 +0000)]
ar71xx: Buffalo WLAE-AG300N initial support

The patch set for Buffalo WLAE-AG300N initial support.

There is another patch for wireless led support that
is posted separately.

Note on serial console:
This unit has buffalo standard 4 pin console, but the unit may not power
on if some console apparatus is connected. This is probably due to some
electronic interaction between the unit's electronic power switch circuit
and the serial console apparatus. If this happens, it is required to power
on the unit without the console, then quickly (re-)attach it.

[juhosg: fix coding style]

Signed-off-by: Yoichi Shinoda <shinoda@jaist.ac.jp>
SVN-Revision: 32931

6 years ago[mcs814x] resync defconfig enable XZ compression
Florian Fainelli [Wed, 1 Aug 2012 12:01:27 +0000 (12:01 +0000)]
[mcs814x] resync defconfig enable XZ compression

SVN-Revision: 32930

6 years ago[package] uclibc++: update to 0.2.4
Florian Fainelli [Wed, 1 Aug 2012 12:01:24 +0000 (12:01 +0000)]
[package] uclibc++: update to 0.2.4

SVN-Revision: 32929

6 years ago[package] uclibc++: stage g++-uc wrappers in $(TOOLCHAIN_DIR)
Florian Fainelli [Wed, 1 Aug 2012 12:01:18 +0000 (12:01 +0000)]
[package] uclibc++: stage g++-uc wrappers in $(TOOLCHAIN_DIR)

Since we are overriding TARGET_CXX with the g++-uc wrapper when
we use cmake it expects to find g++-uc in the TOOLCHAIN_DIR.

SVN-Revision: 32928

6 years ago[mcs814x] properly generate squashfs, jffs2 and initramfs images and kernels
Florian Fainelli [Tue, 31 Jul 2012 23:09:50 +0000 (23:09 +0000)]
[mcs814x] properly generate squashfs, jffs2 and initramfs images and kernels

SVN-Revision: 32927

6 years ago[kernel] disable crashlog on ARM platforms for the moment
Florian Fainelli [Tue, 31 Jul 2012 23:09:46 +0000 (23:09 +0000)]
[kernel] disable crashlog on ARM platforms for the moment

It is currently badly broken on ARM, until fixed, disable it

SVN-Revision: 32926

6 years ago[lantiq] add lantiq svip support
John Crispin [Tue, 31 Jul 2012 18:49:54 +0000 (18:49 +0000)]
[lantiq] add lantiq svip support

SVN-Revision: 32925

6 years agobcm63xx: Add nb6 support (BCM6362 SoC base boards)
Jonas Gorski [Tue, 31 Jul 2012 14:27:47 +0000 (14:27 +0000)]
bcm63xx: Add nb6 support (BCM6362 SoC base boards)

Signed-off-by: Miguel GAIO <miguel.gaio@efixo.com>
Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com>
[jonas.gorski@gmail.com: add network config for nb6, make switch fix a
platform patch]

SVN-Revision: 32924

6 years agobcm63xx: add basic bcm6362 support
Jonas Gorski [Tue, 31 Jul 2012 14:27:42 +0000 (14:27 +0000)]
bcm63xx: add basic bcm6362 support

Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com>
SVN-Revision: 32923

6 years agobcm63xx: enable usb for bcm6328
Jonas Gorski [Tue, 31 Jul 2012 14:27:34 +0000 (14:27 +0000)]
bcm63xx: enable usb for bcm6328

Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com>
SVN-Revision: 32922

6 years agobcm63xx: update enetswitch driver
Jonas Gorski [Tue, 31 Jul 2012 14:27:31 +0000 (14:27 +0000)]
bcm63xx: update enetswitch driver

Update enetswith driver with latest fixes and additions.

Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com>
SVN-Revision: 32921

6 years agobcm63xx: reorder patches
Jonas Gorski [Tue, 31 Jul 2012 14:27:25 +0000 (14:27 +0000)]
bcm63xx: reorder patches

Fix the order and names of patches.

Signed-off-by: Jonas Gorski <jonas.gorski+openwrt@gmail.com>
SVN-Revision: 32920

6 years ago[at91] move i2c module to target modules.mk
Florian Fainelli [Sun, 29 Jul 2012 13:20:51 +0000 (13:20 +0000)]
[at91] move i2c module to target modules.mk

SVN-Revision: 32910

6 years ago[package] partially revert r32855
Florian Fainelli [Sun, 29 Jul 2012 13:20:47 +0000 (13:20 +0000)]
[package] partially revert r32855

Too many modules were removed, while only the octeon module was meant to be
removed, fix that.

SVN-Revision: 32909

6 years ago[omap24xx] resync config
Florian Fainelli [Sun, 29 Jul 2012 13:20:39 +0000 (13:20 +0000)]
[omap24xx] resync config

SVN-Revision: 32908

6 years ago[omap24xx] use the same description as other omap platforms (without the dash)
Florian Fainelli [Sun, 29 Jul 2012 13:20:35 +0000 (13:20 +0000)]
[omap24xx] use the same description as other omap platforms (without the dash)

SVN-Revision: 32907

6 years ago[kernel] refresh pwm patch
John Crispin [Sun, 29 Jul 2012 11:53:06 +0000 (11:53 +0000)]
[kernel] refresh pwm patch

SVN-Revision: 32905

6 years ago[acx-mac80211] dont build on arm as kernel headers are missing/have a different layout
John Crispin [Sun, 29 Jul 2012 09:07:47 +0000 (09:07 +0000)]
[acx-mac80211] dont build on arm as kernel headers are missing/have a different layout

SVN-Revision: 32904

6 years ago[include] fix up scons vs ccache
John Crispin [Sat, 28 Jul 2012 21:06:59 +0000 (21:06 +0000)]
[include] fix up scons vs ccache

SVN-Revision: 32902

6 years ago[kernel] clean up pwm patches and drivers
John Crispin [Sat, 28 Jul 2012 20:47:08 +0000 (20:47 +0000)]
[kernel] clean up pwm patches and drivers

SVN-Revision: 32901

6 years ago[kernel] add config option CONFIG_I2C_MV64XXX to fix kirkwood build
Hauke Mehrtens [Sat, 28 Jul 2012 11:36:52 +0000 (11:36 +0000)]
[kernel] add config option CONFIG_I2C_MV64XXX to fix kirkwood build

SVN-Revision: 32898

6 years ago[realview] resync config
Florian Fainelli [Fri, 27 Jul 2012 15:35:15 +0000 (15:35 +0000)]
[realview] resync config

SVN-Revision: 32895

6 years ago[brcm63xx] fix DVA-G3810BN image generation
Florian Fainelli [Fri, 27 Jul 2012 15:35:12 +0000 (15:35 +0000)]
[brcm63xx] fix DVA-G3810BN image generation

Attached patch fixes a typo in the image Makefile, and overrides the
board name for DVA-G3810BN to set the correct board properties

Signed-Off-By: Jeroen van Bemmel <jbemmel@zonnet.nl>
SVN-Revision: 32894

6 years ago[brcm63xx] add D-link DVA-G3810BN-TL-1 support
Florian Fainelli [Fri, 27 Jul 2012 15:35:09 +0000 (15:35 +0000)]
[brcm63xx] add D-link DVA-G3810BN-TL-1 support

Signed-Off-By: Jeroen van Bemmel <jbemmel@zonnet.nl>
SVN-Revision: 32893

6 years ago[brcm63xx] Add CT6376-1 board support
Florian Fainelli [Fri, 27 Jul 2012 15:35:05 +0000 (15:35 +0000)]
[brcm63xx] Add CT6376-1 board support

Added CT6373 devices instead of using NB4's devices, as Florian suggested.

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 32892

6 years ago[package] base-files: fix detection of standalone ldd
Florian Fainelli [Fri, 27 Jul 2012 15:35:00 +0000 (15:35 +0000)]
[package] base-files: fix detection of standalone ldd

When standalone ldd is selected in config the binary is installed in /usr/bin/ldd.

Signed-off-by: Michael Heimpold <mhei@heimpold.de>
SVN-Revision: 32891

6 years ago[brcm63xx] Add brcmsmac profile
Florian Fainelli [Fri, 27 Jul 2012 15:34:57 +0000 (15:34 +0000)]
[brcm63xx] Add brcmsmac profile

Add brcmsmac profile (kmod-brcmsmac + wpad-mini).

Signed-off-by: Álvaro Fernández Rojas <noltari@gmail.com>
SVN-Revision: 32890

6 years ago[mac80211] mae brcmutil depend on PCI and USB. This broke ar7 fullbuilds
John Crispin [Fri, 27 Jul 2012 08:30:05 +0000 (08:30 +0000)]
[mac80211] mae brcmutil depend on PCI and USB. This broke ar7 fullbuilds

SVN-Revision: 32886

6 years ago[atheros] Add leds back after migration to sysfs
John Crispin [Thu, 26 Jul 2012 17:06:55 +0000 (17:06 +0000)]
[atheros] Add leds back after migration to sysfs

atheros trunk moved to full sysfs gpiolib, but the leds were forgotten.
This restores the wlan led that was missing after switching from backfire
to trunk.

Signed-off-by: Karl Palsson <karlp@remake.is>
SVN-Revision: 32885

6 years ago[brcm2708] add more missing config symbols
Florian Fainelli [Thu, 26 Jul 2012 17:03:08 +0000 (17:03 +0000)]
[brcm2708] add more missing config symbols

SVN-Revision: 32884

6 years ago[generic] add more USB-related config symbols
Florian Fainelli [Thu, 26 Jul 2012 17:03:05 +0000 (17:03 +0000)]
[generic] add more USB-related config symbols

SVN-Revision: 32883

6 years ago[ep93xx] make sure target modules depend on the target
Florian Fainelli [Thu, 26 Jul 2012 17:03:02 +0000 (17:03 +0000)]
[ep93xx] make sure target modules depend on the target

SVN-Revision: 32882

6 years ago[kernel] bcma: update to version from wireless-testing tag master-2012-07-16
Hauke Mehrtens [Thu, 26 Jul 2012 15:42:18 +0000 (15:42 +0000)]
[kernel] bcma: update to version from wireless-testing tag master-2012-07-16

I did not add the last patch, because it add some stubs for the flash drivers and our patches have to be changed for that, if the flash support in bcma is able to do something I will integrate it.

This also contains 3 additional patches fixing some bugs in bcma.

SVN-Revision: 32880

6 years ago[package] make kmod-mvsas depend on x86 since it fails building on other targets...
Florian Fainelli [Thu, 26 Jul 2012 09:29:22 +0000 (09:29 +0000)]
[package] make kmod-mvsas depend on x86 since it fails building on other targets (eg: brcm63xx)

SVN-Revision: 32878

6 years ago[ar71xx] fix typo in modules.mk
John Crispin [Wed, 25 Jul 2012 19:52:30 +0000 (19:52 +0000)]
[ar71xx] fix typo in modules.mk

This patch fixes a small typo in kmod-leds-wndr3700-usb description, where the board is for some reason made by NETGWR, not NETGEAR.

Signed-off-by: Roman A. aka BasicXP <x12ozmouse@ya.ru>
SVN-Revision: 32870

6 years ago[package] make kmod-fb dependent on DISPLAY_SUPPORT and update targets accordingly
Florian Fainelli [Wed, 25 Jul 2012 19:36:19 +0000 (19:36 +0000)]
[package] make kmod-fb dependent on DISPLAY_SUPPORT and update targets accordingly

SVN-Revision: 32869

6 years agomtd: fix trx_fixup
Felix Fietkau [Wed, 25 Jul 2012 17:28:32 +0000 (17:28 +0000)]
mtd: fix trx_fixup

With the BRCM47xx image I have built (Linksys E3000), there are additional
writes following the completion of "mtd_replace_jffs2" which invalidate the
'trx_fixup' performed by 'mtd_replace_jffs2'.  Moving the 'trx_fixup' to somewhere
after all writes have completed fixes the problem.  I also noticed that 'erasesize'
used to compute 'block_offset' in 'mtd_fixtrx' is used before it is computed by
'mtd_check_open'; moving the call to 'mtd_check_open' up a few lines fixes this.

Unlike 'mtd_fixtrx', 'trx_fixup' appears to assume that the TRX header is always at
offset 0; which may be the cause of the problem described in Ticket #8960.

Signed-off-by: Nathan Hintz <nlhintz@hotmail.com>
SVN-Revision: 32866

6 years ago[ppc40x] move isp116x driver to target specific modules.mk
Florian Fainelli [Wed, 25 Jul 2012 15:50:45 +0000 (15:50 +0000)]
[ppc40x] move isp116x driver to target specific modules.mk

SVN-Revision: 32865

6 years ago[ramips] move rt305x-dwc driver to target specific modules.mk
Florian Fainelli [Wed, 25 Jul 2012 15:50:42 +0000 (15:50 +0000)]
[ramips] move rt305x-dwc driver to target specific modules.mk

SVN-Revision: 32864

6 years ago[brcm47xx] move ubsec-ssb driver to target specific modules.mk
Florian Fainelli [Wed, 25 Jul 2012 15:50:38 +0000 (15:50 +0000)]
[brcm47xx] move ubsec-ssb driver to target specific modules.mk

SVN-Revision: 32863