openwrt/svn-archive/archive.git
9 years agoubox: update to latest git HEAD
Rafał Miłecki [Tue, 14 Jul 2015 10:32:51 +0000 (10:32 +0000)]
ubox: update to latest git HEAD

Signed-off-by: John Crispin <blogic@openwrt.org>
This adds support for filtering messages
Backport of r46356

SVN-Revision: 46365

9 years agoodhcp6c: fix some more compatibility issues
Steven Barth [Mon, 13 Jul 2015 21:52:25 +0000 (21:52 +0000)]
odhcp6c: fix some more compatibility issues

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 46337

9 years agoodhcp6c: work-around more ISP DHCPv6-PD issues
Steven Barth [Mon, 13 Jul 2015 21:52:14 +0000 (21:52 +0000)]
odhcp6c: work-around more ISP DHCPv6-PD issues

Signed-off-by: Steven Barth <steven@midlink.org>
SVN-Revision: 46336

9 years agoCC: curl: fix some security vulnerabilities
Hauke Mehrtens [Sun, 12 Jul 2015 16:03:51 +0000 (16:03 +0000)]
CC: curl: fix some security vulnerabilities
This fixes the following security vulnerabilities in curl:
* CVE-2015-3143
* CVE-2015-3144
* CVE-2015-3145
* CVE-2015-3148
* CVE-2015-3153
* CVE-2015-3236
* CVE-2015-3237

This was fixed in trunk with update to version 7.43.0 in r46169.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46312

9 years agoCC: kernel: fix loading bluetooth_6lowpan
Hauke Mehrtens [Sun, 12 Jul 2015 16:03:09 +0000 (16:03 +0000)]
CC: kernel: fix loading bluetooth_6lowpan
Backprot of r46179

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46311

9 years agoCC: kernel: add missing config option
Hauke Mehrtens [Sun, 12 Jul 2015 16:02:47 +0000 (16:02 +0000)]
CC: kernel: add missing config option
This should fix build of sunxi.

Backport of r46304

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46310

9 years agoCC: kernel: update 3.18 to 3.18.18
Hauke Mehrtens [Sun, 12 Jul 2015 16:02:34 +0000 (16:02 +0000)]
CC: kernel: update 3.18 to 3.18.18
Changelog:
 * https://www.kernel.org/pub/linux/kernel/v3.x/ChangeLog-3.18.18

Backport of r46299, r46303 and r46308.

Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
SVN-Revision: 46309

9 years agoebtables: remove partially broken musl fix
Felix Fietkau [Fri, 10 Jul 2015 11:37:36 +0000 (11:37 +0000)]
ebtables: remove partially broken musl fix

See http://patchwork.ozlabs.org/patch/493573/ for more information

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 46293

9 years agotoolchain: mark musl as broken for the chaos_calmer branch
Felix Fietkau [Fri, 10 Jul 2015 11:28:44 +0000 (11:28 +0000)]
toolchain: mark musl as broken for the chaos_calmer branch

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 46291

9 years agomac80211: add brcmfmac patch fixing user space picked MACs
Rafał Miłecki [Thu, 9 Jul 2015 18:23:41 +0000 (18:23 +0000)]
mac80211: add brcmfmac patch fixing user space picked MACs

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46288

SVN-Revision: 46290

9 years agomac80211: try to workaround some brcmfmac bug with handling interfaces
Rafał Miłecki [Thu, 9 Jul 2015 18:23:35 +0000 (18:23 +0000)]
mac80211: try to workaround some brcmfmac bug with handling interfaces

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46278

SVN-Revision: 46289

9 years agoCC: openssl: update to v1.0.2d (CVE-2015-1793)
Jo-Philipp Wich [Thu, 9 Jul 2015 13:15:19 +0000 (13:15 +0000)]
CC: openssl: update to v1.0.2d (CVE-2015-1793)
During certificate verification, OpenSSL (starting from version 1.0.1n and
1.0.2b) will attempt to find an alternative certificate chain if the first
attempt to build such a chain fails. An error in the implementation of this
logic can mean that an attacker could cause certain checks on untrusted
certificates to be bypassed, such as the CA flag, enabling them to use a valid
leaf certificate to act as a CA and "issue" an invalid certificate.

This issue will impact any application that verifies certificates including
SSL/TLS/DTLS clients and SSL/TLS/DTLS servers using client authentication.

Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Backport of r46285

SVN-Revision: 46286

9 years agox86/xen_domu: enable image generation
John Crispin [Thu, 9 Jul 2015 05:29:25 +0000 (05:29 +0000)]
x86/xen_domu: enable image generation

Add features ext4 targz to target x86/xen_domu in order to
generate images in defconfig.

This fixes #18074.

Signed-off-by: Luiz Angelo Daros de Luca <luizluca@gmail.com>
Backport of r46233

SVN-Revision: 46277

9 years agonvram: get rid of NVRAM_SPACE
Rafał Miłecki [Wed, 8 Jul 2015 06:06:07 +0000 (06:06 +0000)]
nvram: get rid of NVRAM_SPACE

Now we support NVRAM whatever its size is.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46262

SVN-Revision: 46263

9 years agonvram: fix displayed info about NVRAM size
Rafał Miłecki [Wed, 8 Jul 2015 05:45:09 +0000 (05:45 +0000)]
nvram: fix displayed info about NVRAM size

Use actual partition size and content offset to calculate it.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46260

SVN-Revision: 46261

9 years agoRevert "Revert "nvram: increase NVRAM size to 64 KiB""
Rafał Miłecki [Wed, 8 Jul 2015 04:55:10 +0000 (04:55 +0000)]
Revert "Revert "nvram: increase NVRAM size to 64 KiB""

This reverts commit r45579.

With the latest change r46251 ("nvram: fix "Segmentation fault" caused
by setting memory out of buffer") nvram utility shouldn't crash anymore.

It was tested on 3 brcm47xx devices:
1) Unknown with 0x10000 NVRAM size (0x8000 offset)
2) Linksys E1000 V2.1 with 0x10000 (0x8000 offset)
3) Linksys WRT300N V1 with 0x10000 (0x8000 offset)

And 3 bcm53xx devices:
1) Buffalo WZR-600DHP2 with 0x160000 NVRAM size
2) Buffalo WZR-1750DHP with 0x10000 NVRAM size
3) Netgear R6250 V1 with 0x180000 NVRAM size
(all using 0 offset)

This is an important change as it allows reading whole NVRAM. This may
critical when reading some basic configuration (e.g. switch ports).

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46253

SVN-Revision: 46259

9 years agonvram: fix "Segmentation fault" caused by setting memory out of buffer
Rafał Miłecki [Wed, 8 Jul 2015 04:55:04 +0000 (04:55 +0000)]
nvram: fix "Segmentation fault" caused by setting memory out of buffer

Some MTD partitions with NVRAM have content starting in the middle. In
such case offset is set and nvram_header returns pointer to the middle.
It means we have to respect offset when calculating remaining space.

By the way use real MTD partition size (nvram_part_size variable) as we
may want to bump NVRAM_SPACE in the (very near) future.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46251

SVN-Revision: 46258

9 years agoprocd: update to the latest version, fix stdout logging
Rafał Miłecki [Wed, 8 Jul 2015 04:54:57 +0000 (04:54 +0000)]
procd: update to the latest version, fix stdout logging

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46212

SVN-Revision: 46257

9 years agoar71xx: fix kernel Oops in at803x_link_change_notify
Felix Fietkau [Tue, 7 Jul 2015 21:52:29 +0000 (21:52 +0000)]
ar71xx: fix kernel Oops in at803x_link_change_notify

r45954 ("ar71xx: fix 100/10mbps ethernet link issues on mynet range
extender") introduced a pdata based modification of the tx_clk_dly. But it
was not checked if pdata actually existed. This caused a page fault on all
devices which didn't have at803x_platform_data specified for an at803x
based device.

Signed-off-by: Sven Eckelmann <sven@open-mesh.com>
Backport of r46200

SVN-Revision: 46252

9 years agoscripts/feeds: observe -p flag for preferential feeds
Steven Barth [Tue, 7 Jul 2015 11:30:39 +0000 (11:30 +0000)]
scripts/feeds: observe -p flag for preferential feeds

lookup_target was trampling the $feed variable, resulting in the -p flag
no longer preferentially installing from the named feed.

Make sure to use a local variable for this instead.

Signed-off-by: Karl Palsson <karlp@remake.is>
SVN-Revision: 46213

9 years agoCC: kernel: kmod-scsi-core: fix load on install
Jonas Gorski [Sat, 4 Jul 2015 12:53:57 +0000 (12:53 +0000)]
CC: kernel: kmod-scsi-core: fix load on install
sd_mod depends on scsi_mod, but due to it being an AutoLoad and not
AutoProbe module, it was not loading when installing the package,
causing unknown symbol errors for sd_mod and anything depending on it.

Closes #14927, #18293, #19351.

Backport of r46176.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46177

9 years agoCC: uclibc++: make g++-uc* wrappers relocatable
Jo-Philipp Wich [Fri, 3 Jul 2015 13:38:12 +0000 (13:38 +0000)]
CC: uclibc++: make g++-uc* wrappers relocatable
The g++-uc wrapper hardcodes $(STAGING_DIR) and $(TOOLCHAIN_DIR) paths which
will not work outside of the original build environment.

Replace the hardcoded staging_dir occurences with paths relative to the
$STAGING_DIR environment variable to make the g++-uc* wrappers usable in an
SDK environment.

Fixes the libdb47 build failure reported at
  https://lists.openwrt.org/pipermail/openwrt-devel/2015-April/032455.html

Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Backport of r46162

SVN-Revision: 46163

9 years agoath9k: make DMA stop related messages debug-only (#11862, #9654)
Felix Fietkau [Thu, 2 Jul 2015 11:45:39 +0000 (11:45 +0000)]
ath9k: make DMA stop related messages debug-only (#11862, #9654)

A long time ago, ath9k had issues during reset where the DMA engine
would stay active and could potentially corrupt memory.
To debug those issues, the driver would print warnings whenever they
occur.

Nowadays, these issues are gone and the primary cause of these messages
is if the MAC is stuck during reset or busy processing a long
transmission. This is fairly harmless, yet these messages continue to
worry users.

To reduce the number of bogus bug reports, turn these messages into
debug messages and count their occurence in the "reset" debugfs file.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r46158

SVN-Revision: 46159

9 years agomdns: fix ubus wait_for command
Felix Fietkau [Thu, 2 Jul 2015 11:26:28 +0000 (11:26 +0000)]
mdns: fix ubus wait_for command

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r46156

SVN-Revision: 46157

9 years agoCC: kernel: update 3.18 to 3.18.17
Jonas Gorski [Wed, 1 Jul 2015 00:32:10 +0000 (00:32 +0000)]
CC: kernel: update 3.18 to 3.18.17
Changelog:
* https://www.kernel.org/pub/linux/kernel/v3.x/ChangeLog-3.18.17

Backport of r46150.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46155

9 years agoCC: brcm63xx: swap a4001n power leds again
Jonas Gorski [Wed, 1 Jul 2015 00:32:03 +0000 (00:32 +0000)]
CC: brcm63xx: swap a4001n power leds again
This partially reverts r44577 ("brcm63xx: fix DTS LEDs and buttons").

Fixes #19858.

Backport of r46149.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46154

9 years agoCC: brcm63xx: unbreak board_probe_switch again
Jonas Gorski [Wed, 1 Jul 2015 00:32:01 +0000 (00:32 +0000)]
CC: brcm63xx: unbreak board_probe_switch again
Also move it to an earlier place so new boards added are less likely
to confuse quilt.

Fixes #19965.

Backport of r46148.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
SVN-Revision: 46153

9 years agoRevert "ar71xx: add profile and build image for EnGenius EPG5000."
John Crispin [Mon, 29 Jun 2015 08:24:13 +0000 (08:24 +0000)]
Revert "ar71xx: add profile and build image for EnGenius EPG5000."

This reverts commit ac3184da4138c54b8a2c1be637be7cf78fc40fce.

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 46144

9 years agoRevert "ar71xx: add user-space support for EnGenius EPG5000."
John Crispin [Mon, 29 Jun 2015 08:24:07 +0000 (08:24 +0000)]
Revert "ar71xx: add user-space support for EnGenius EPG5000."

This reverts commit e3326c4473bc6b7b5fd899ac955af974ebc64b92.

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 46143

9 years agoRevert "ar71xx: add kernel support for EnGenius EPG5000."
John Crispin [Mon, 29 Jun 2015 08:24:03 +0000 (08:24 +0000)]
Revert "ar71xx: add kernel support for EnGenius EPG5000."

This reverts commit 27386b74adfe868f693176f55861984432afd2ce.

Signed-off-by: John Crispin <blogic@openwrt.org>
SVN-Revision: 46142

9 years agob53: Allow using all 8 ports on BCM53011
Rafał Miłecki [Mon, 22 Jun 2015 07:59:09 +0000 (07:59 +0000)]
b53: Allow using all 8 ports on BCM53011

On two tested devices: Netgear R6250 (BCM53011 rev 2) and Luxul XWC-1000
(BCM53011 rev 3) it was possible to use port 7 and eth1 (instead of port
5 and eth0). It seems BCM53011 just like BCM53012 has 8 ports and
usually 3 of them are connected to the SoC.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46104

SVN-Revision: 46107

9 years agobcm53xx: respect ACTIVE_LOW when powering USB using GPIO
Rafał Miłecki [Sun, 21 Jun 2015 20:08:32 +0000 (20:08 +0000)]
bcm53xx: respect ACTIVE_LOW when powering USB using GPIO

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46088

SVN-Revision: 46103

9 years agobcm53xx: enable USB power on Buffalo WZR-1750DHP
Rafał Miłecki [Sun, 21 Jun 2015 20:08:27 +0000 (20:08 +0000)]
bcm53xx: enable USB power on Buffalo WZR-1750DHP

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46087

SVN-Revision: 46102

9 years agobcm53xx: update SPROM driver to include just-sent patch
Rafał Miłecki [Sun, 21 Jun 2015 20:08:21 +0000 (20:08 +0000)]
bcm53xx: update SPROM driver to include just-sent patch

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46086

SVN-Revision: 46101

9 years agomac80211: backport some brcmfmac patches
Rafał Miłecki [Sun, 21 Jun 2015 20:08:11 +0000 (20:08 +0000)]
mac80211: backport some brcmfmac patches

There are two important patches in this patchset: updating read pointer
quicker & rework of .get_station().
There are few more upstream patches that are p2p-related and weren't
backported in this commit.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46084

SVN-Revision: 46100

9 years agobcm53xx: enable USB power on Netgear R6250
Rafał Miłecki [Sun, 21 Jun 2015 20:08:03 +0000 (20:08 +0000)]
bcm53xx: enable USB power on Netgear R6250

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46083

SVN-Revision: 46099

9 years agobcm53xx: update xHCI patch
Rafał Miłecki [Sun, 21 Jun 2015 20:07:58 +0000 (20:07 +0000)]
bcm53xx: update xHCI patch

This drops some debugging pr_info and adds platform_device_unregister.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46082

SVN-Revision: 46098

9 years agobcm53xx: fix enabling GPIO-controlled USB power
Rafał Miłecki [Sun, 21 Jun 2015 20:07:52 +0000 (20:07 +0000)]
bcm53xx: fix enabling GPIO-controlled USB power

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46081

SVN-Revision: 46097

9 years agobcm53xx: add xHCI support
Rafał Miłecki [Sun, 21 Jun 2015 20:07:44 +0000 (20:07 +0000)]
bcm53xx: add xHCI support

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46080

SVN-Revision: 46096

9 years agoramips: refresh patch with USB_MT7621_XHCI_PLATFORM
Rafał Miłecki [Sun, 21 Jun 2015 20:06:31 +0000 (20:06 +0000)]
ramips: refresh patch with USB_MT7621_XHCI_PLATFORM

Fixes: edea6dfde4dc ("kernel: make USB_XHCI_PLATFORM selectable")
Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46085

SVN-Revision: 46095

9 years agokernel: make USB_XHCI_PLATFORM selectable
Rafał Miłecki [Sun, 21 Jun 2015 20:06:24 +0000 (20:06 +0000)]
kernel: make USB_XHCI_PLATFORM selectable

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46079

SVN-Revision: 46094

9 years agootrx: optimize memory usage by function checking TRX CRC
Rafał Miłecki [Sun, 21 Jun 2015 20:04:48 +0000 (20:04 +0000)]
otrx: optimize memory usage by function checking TRX CRC

Allocating huge buffer for the whole TRX wasn't too optimal.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46078

SVN-Revision: 46093

9 years agootrx: make crc32 table global for further optimizations
Rafał Miłecki [Sun, 21 Jun 2015 20:04:42 +0000 (20:04 +0000)]
otrx: make crc32 table global for further optimizations

So far we got only one generic function accessing this table, but
implementing optimizations will require calculating crc32 in other code
parts as well.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r46077

SVN-Revision: 46092

9 years agoath9k: fix ar956x device detection
Felix Fietkau [Sun, 21 Jun 2015 17:52:47 +0000 (17:52 +0000)]
ath9k: fix ar956x device detection

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r46090

SVN-Revision: 46091

9 years agomwlwifi: upgrade to 10.3.0.3 (backport of r46061)
Imre Kaloz [Fri, 19 Jun 2015 08:35:57 +0000 (08:35 +0000)]
mwlwifi: upgrade to 10.3.0.3 (backport of r46061)

Signed-off-by: Imre Kaloz <kaloz@openwrt.org>
SVN-Revision: 46062

9 years agoralink: add support for mt7621 switch counters
John Crispin [Thu, 18 Jun 2015 20:24:49 +0000 (20:24 +0000)]
ralink: add support for mt7621 switch counters

Signed-off-by: Cristian Morales Vega <cristian@samknows.com>
Backport of r46050

SVN-Revision: 46060

9 years agoar71xx: add profile and build image for EnGenius EPG5000.
John Crispin [Thu, 18 Jun 2015 20:24:45 +0000 (20:24 +0000)]
ar71xx: add profile and build image for EnGenius EPG5000.

Signed-off-by: Christian Beier <cb@shoutrlabs.com>
Backport of r46049

SVN-Revision: 46059

9 years agoar71xx: add user-space support for EnGenius EPG5000.
John Crispin [Thu, 18 Jun 2015 20:24:41 +0000 (20:24 +0000)]
ar71xx: add user-space support for EnGenius EPG5000.

Signed-off-by: Christian Beier <cb@shoutrlabs.com>
Backport of r46048

SVN-Revision: 46058

9 years agoar71xx: add kernel support for EnGenius EPG5000.
John Crispin [Thu, 18 Jun 2015 20:24:31 +0000 (20:24 +0000)]
ar71xx: add kernel support for EnGenius EPG5000.

Signed-off-by: Christian Beier <cb@shoutrlabs.com>
Backport of r46047

SVN-Revision: 46057

9 years agoar71xx: add support for TL-WR720N-v4.
John Crispin [Thu, 18 Jun 2015 20:24:26 +0000 (20:24 +0000)]
ar71xx: add support for TL-WR720N-v4.

The hardware should be almost the same as TL-WR720N-v3.  WiFi and LAN networks
were tested by "Lo Yuk Fai <loyukfai@gmail.com>".  Failsafe and slider switch
were tested by "Wong min <alpha080@gmail.com>".

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
Backport of r46046

SVN-Revision: 46056

9 years agoar71xx: add support for compex wpj531
John Crispin [Thu, 18 Jun 2015 20:24:20 +0000 (20:24 +0000)]
ar71xx: add support for compex wpj531

tftpboot 0x80500000 openwrt-ar71xx-generic-wpj531-16M-squashfs-sysupgrade.bin
erase 0x9f030000 +$filesize
erase 0x9f680000 +1
cp.b $fileaddr 0x9f030000 $filesize

Signed-off-by: Christian Mehlis <christian@m3hlis.de>
Backport of r46045

SVN-Revision: 46055

9 years agoar71xx: Add Support for the Bitmain Antminer S1
John Crispin [Thu, 18 Jun 2015 20:24:15 +0000 (20:24 +0000)]
ar71xx: Add Support for the Bitmain Antminer S1

This patch adds support for the Bitmain Antminer S1 Cryptocurrency Miner

http://wiki.openwrt.org/toh/bitmain/s1

Signed-off-by: L. D. Pinney <ldpinney@gmail.com>
Acked-by: James Hilliard <james.hilliard1@gmail.com>
Backport of r46044

SVN-Revision: 46054

9 years agoramips: Add Support for the Linksys RE6500
John Crispin [Thu, 18 Jun 2015 20:24:12 +0000 (20:24 +0000)]
ramips: Add Support for the Linksys RE6500

This patch adds support for the Linksys RE6500 Range Extender

http://wiki.openwrt.org/toh/linksys/re6500

Signed-off-by: L. D. Pinney <ldpinney@gmail.com>
Backport of r46043

SVN-Revision: 46053

9 years agoramips: Add support for the D-Link DCS-930 B1
John Crispin [Thu, 18 Jun 2015 20:24:09 +0000 (20:24 +0000)]
ramips: Add support for the D-Link DCS-930 B1

http://wiki.openwrt.org/toh/d-link/dcs-930l

Signed-off-by: L. D. Pinney <ldpinney@gmail.com>
Backport of r46042

SVN-Revision: 46052

9 years agoramips: add support for PandoraBox PBR-M1
John Crispin [Thu, 18 Jun 2015 20:24:05 +0000 (20:24 +0000)]
ramips: add support for PandoraBox PBR-M1

This patch is for PandoraBox PBR-M1 which is based on mt7621,
all the features work fine, including rtc, leds, button, usb3.0, etc.

Signed-off-by: tymon <banglang.huang@foxmail.com>
Backport of r46041

SVN-Revision: 46051

9 years agoramips: fix a bug in the cevt code on mt7620
John Crispin [Thu, 18 Jun 2015 12:06:15 +0000 (12:06 +0000)]
ramips: fix a bug in the cevt code on mt7620

Signed-off-by: John Crispin <blogic@openwrt.org>
Backport of r46038

SVN-Revision: 46039

9 years agoar71xx: fix hw id of tl-wr710n image (#19902)
Felix Fietkau [Thu, 18 Jun 2015 09:42:00 +0000 (09:42 +0000)]
ar71xx: fix hw id of tl-wr710n image (#19902)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r46032

SVN-Revision: 46033

9 years agoprocd: update to the latest version, fixes umask handling for creating /dev/shm
Felix Fietkau [Thu, 18 Jun 2015 06:42:11 +0000 (06:42 +0000)]
procd: update to the latest version, fixes umask handling for creating /dev/shm

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r46023

SVN-Revision: 46028

9 years agoCC: x86/64: enable VirtIO block and console support
Jo-Philipp Wich [Wed, 17 Jun 2015 11:36:51 +0000 (11:36 +0000)]
CC: x86/64: enable VirtIO block and console support
This is required to use x86/64 in KVM containers with VirtIO disks

Signed-off-by: Jo-Philipp Wich <jow@openwrt.org>
Backport of r46017

SVN-Revision: 46018

9 years agoar71xx: rb493g: fix gpio-mdio accesses under load
Jonas Gorski [Wed, 17 Jun 2015 10:22:06 +0000 (10:22 +0000)]
ar71xx: rb493g: fix gpio-mdio accesses under load

It seems that there are maximum timings for mdio accesses that can be
hit when the system is under load and the thread is scheduled during
a read or write access. Since there is no way of knowing if this
happens as there is not even a parity bit, try to work around it
by disabling interrupts during any gpio-mdio accesses.

Performance impact seems to be neglectable, as mdio accesses are not
that often.

Hopefully fixes #19500.

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Backport of r46012.

SVN-Revision: 46016

9 years agokernel: update 3.18 to 3.18.16
Jonas Gorski [Wed, 17 Jun 2015 10:22:04 +0000 (10:22 +0000)]
kernel: update 3.18 to 3.18.16

Changelogs:

* https://www.kernel.org/pub/linux/kernel/v3.x/ChangeLog-3.18.15
* https://www.kernel.org/pub/linux/kernel/v3.x/ChangeLog-3.18.16

Signed-off-by: Jonas Gorski <jogo@openwrt.org>
Backport of r46011.

SVN-Revision: 46015

9 years agoprocd: update to the latest version, fixes /dev/shm permissions
Felix Fietkau [Wed, 17 Jun 2015 10:13:42 +0000 (10:13 +0000)]
procd: update to the latest version, fixes /dev/shm permissions

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r46013

SVN-Revision: 46014

9 years agoath9k: add another fix for some rx/tx dma stop issues, refresh patches
Felix Fietkau [Wed, 17 Jun 2015 09:11:08 +0000 (09:11 +0000)]
ath9k: add another fix for some rx/tx dma stop issues, refresh patches

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r46008

SVN-Revision: 46010

9 years agokernel: add missing config symbol
Felix Fietkau [Wed, 17 Jun 2015 09:10:08 +0000 (09:10 +0000)]
kernel: add missing config symbol

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r46007

SVN-Revision: 46009

9 years agobase-files: replace snapshot key with 15.05 release key
Felix Fietkau [Tue, 16 Jun 2015 13:38:37 +0000 (13:38 +0000)]
base-files: replace snapshot key with 15.05 release key

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 46002

9 years agoopkg: fix duplicate check_signature line for smime variant (#19863)
Felix Fietkau [Tue, 16 Jun 2015 10:51:38 +0000 (10:51 +0000)]
opkg: fix duplicate check_signature line for smime variant (#19863)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r45999

SVN-Revision: 46000

9 years agobcm53xx: use USB patches sent upstream by Hauke
Rafał Miłecki [Tue, 16 Jun 2015 08:32:17 +0000 (08:32 +0000)]
bcm53xx: use USB patches sent upstream by Hauke

This stabilizes USB support. The old patch was handling initialization
in a different order that was causing some problems with few USB 3.0
devices. Some weren't detected, some were working unstable, sometimes
USB 3.0 could hang the whole controller.

A still known issue (but not a regression) is controller hang triggered
by connecting USB 1.1 device when not having OHCI controller enabled
(kmod-usb-ohci).

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r45997

SVN-Revision: 45998

9 years agokernel: use bcma-hcd.ko on bcm53xx for ohci
Rafał Miłecki [Tue, 16 Jun 2015 05:04:53 +0000 (05:04 +0000)]
kernel: use bcma-hcd.ko on bcm53xx for ohci

On bcm53xx we still need usb-ohci to support USB 1.1 devices and it
obviously needs bcma-hcd.ko that will init controller and register
platform device.

Signed-off-by: Rafał Miłecki <zajec5@gmail.com>
Backport of r45994

SVN-Revision: 45996

9 years agokernel: add missing config symbol (#19870)
Felix Fietkau [Mon, 15 Jun 2015 21:16:55 +0000 (21:16 +0000)]
kernel: add missing config symbol (#19870)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r45992

SVN-Revision: 45993

9 years agomt76: update to the latest version to fix a build error (#19871)
Felix Fietkau [Mon, 15 Jun 2015 21:15:36 +0000 (21:15 +0000)]
mt76: update to the latest version to fix a build error (#19871)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r45990

SVN-Revision: 45991

9 years agoversion.mk: adjust for CC release
Felix Fietkau [Mon, 15 Jun 2015 13:42:43 +0000 (13:42 +0000)]
version.mk: adjust for CC release

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 45989

9 years agoUpdate dnsmasq to v2.73.
Steven Barth [Mon, 15 Jun 2015 08:00:46 +0000 (08:00 +0000)]
Update dnsmasq to v2.73.

Signed-off-by: Toke Høiland-Jørgensen <toke@toke.dk>
SVN-Revision: 45987

9 years agoRevert "kernel: add rtl8723au driver (staging)"
Felix Fietkau [Mon, 15 Jun 2015 07:46:55 +0000 (07:46 +0000)]
Revert "kernel: add rtl8723au driver (staging)"

This one cannot possibly work, as it depends on cfg80211, which is built
out of tree.

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
Backport of r45985

SVN-Revision: 45986

9 years agofeeds.conf.default: use the for-15.05 branches, drop targets feed
Felix Fietkau [Sun, 14 Jun 2015 18:35:59 +0000 (18:35 +0000)]
feeds.conf.default: use the for-15.05 branches, drop targets feed

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 45978

9 years agoremove broken targets
Felix Fietkau [Sun, 14 Jun 2015 18:17:54 +0000 (18:17 +0000)]
remove broken targets

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 45977

9 years agogcc: remove version 4.9
Felix Fietkau [Sun, 14 Jun 2015 18:17:49 +0000 (18:17 +0000)]
gcc: remove version 4.9

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 45976

9 years agokernel: remove linux 4.0
Felix Fietkau [Sun, 14 Jun 2015 18:17:47 +0000 (18:17 +0000)]
kernel: remove linux 4.0

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 45975

9 years agoadd chaos_calmer branch
Felix Fietkau [Sun, 14 Jun 2015 17:57:16 +0000 (17:57 +0000)]
add chaos_calmer branch

SVN-Revision: 45974

9 years agoMakefile: move the cleaning of staging_dir/target* from dirclean to clean
Felix Fietkau [Sun, 14 Jun 2015 17:47:16 +0000 (17:47 +0000)]
Makefile: move the cleaning of staging_dir/target* from dirclean to clean

Currently "make clean" only clears the build_dir/target*, but leaves
staging_dir/target* intact. "make clean" should also clean the
staging_dir/target* directories, as in the current situation some old
packages or libraries may be linked into the firmware from staging_dir
despite a "make clean".

The patch reorganises clean / dirclean functionality so that
* "make clean" also clears the staging_dir/target* in addition to
build_dir/target*.
* "make dirclean" clears toolchain and host(=tools) directories from both
build_dir and staging_dir

signed-off-by: Hannu Nyman <hannu.nyman@iki.fi>

SVN-Revision: 45973

9 years agoprocd: update to the latest version, adds a few fixes
Felix Fietkau [Sun, 14 Jun 2015 17:44:04 +0000 (17:44 +0000)]
procd: update to the latest version, adds a few fixes

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 45972

9 years agoar71xx: return limit as number of processed frames when re-scheduling NAPI poll
Felix Fietkau [Sun, 14 Jun 2015 17:43:59 +0000 (17:43 +0000)]
ar71xx: return limit as number of processed frames when re-scheduling NAPI poll

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 45971

9 years agoar8216: Fix problem with AR8337 MAC swap handling
Felix Fietkau [Sun, 14 Jun 2015 17:43:50 +0000 (17:43 +0000)]
ar8216: Fix problem with AR8337 MAC swap handling

AR8337 supports a configuration bit to swap MAC0 and MAC6.
Currently this is set in general if an AR8337 is detected and causes
issues with devices using an AR8334 (internally an AR8337, just
less chip pins).
And it might even cause issues with AR8337-based devices with
different board designs.

Swapping the MAC's however isn't needed for AR8337 in general.
It's just needed in case of certain board designs (affected devices
seem to be based on Atheros reference board AP135/136-010).
Therefore this configuration bit should be moved to platform data.

The patch includes the needed changes to the device initialization
code of affected devices. Hopefully I didn't miss any ..

Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
SVN-Revision: 45970

9 years agoelfutils: import package from packages.git
Felix Fietkau [Sun, 14 Jun 2015 17:43:40 +0000 (17:43 +0000)]
elfutils: import package from packages.git

elfutils is required by perf. So we'll move this package from
packages.git and make it part of the core distribution.

Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 45969

9 years agobzip2: import package from packages.git
Felix Fietkau [Sun, 14 Jun 2015 17:43:34 +0000 (17:43 +0000)]
bzip2: import package from packages.git

bzip2 is required by elfutils, itself required by perf. So we'll move
this package from packages.git and make it part of the core
distribution.

Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 45968

9 years agoargp-standalone: import package from packages.git
Felix Fietkau [Sun, 14 Jun 2015 17:43:28 +0000 (17:43 +0000)]
argp-standalone: import package from packages.git

argp-standalone is required by elfutils, itself required by perf. So
we'll move this package from packages.git and make it part of the core
distribution.

Signed-off-by: Mathieu Olivari <mathieu@codeaurora.org>
SVN-Revision: 45967

9 years agoinclude/scan.mk: Add -a argument to the grep to treat all Makefiles as a text file.
Felix Fietkau [Sun, 14 Jun 2015 17:43:22 +0000 (17:43 +0000)]
include/scan.mk: Add -a argument to the grep to treat all Makefiles as a text file.

Hello,

I found out that in some rare cases grep can treat Makefile as a binary file. That happened to me on UTF-8 Gentoo if Makefile contained a character which was from extended ASCII table.
Without this patch the output of the grep command in the $(FILELIST) target is not valid (contains line "Binary file matches") and following commands fail to create feed index file.

Best Regards,
Martin Strbacka

Signed-off-by: Martin Strbacka <martin.strbacka@nic.cz>
SVN-Revision: 45966

9 years agoar71xx: Add pll_1000 value for eth0 to Ubiquiti Rocket M XW
Felix Fietkau [Sun, 14 Jun 2015 17:43:18 +0000 (17:43 +0000)]
ar71xx: Add pll_1000 value for eth0 to Ubiquiti Rocket M XW

This patch adds the pll_1000 value for eth0 interface. This makes the Rocket M
XW image compatible with other Ubiquiti devices with similar hardware with a
Gigabit Ethernet port.

Signed-off-by: Roger Pueyo Centelles <roger.pueyo@guifi.net>
SVN-Revision: 45965

9 years agobuild: do another init script enabling run, fixes init scripts added via files/ ...
Felix Fietkau [Sun, 14 Jun 2015 17:42:05 +0000 (17:42 +0000)]
build: do another init script enabling run, fixes init scripts added via files/ (#19857)

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 45964

9 years agooxnas: change file extension of u-boot+initramfs image
Felix Fietkau [Sun, 14 Jun 2015 17:42:00 +0000 (17:42 +0000)]
oxnas: change file extension of u-boot+initramfs image

The *u-boot-initramfs* images are not ITB/FIT images but are rather
supposed to be loaded to RAM and jumped-into.
To avoid confusion, change the file extension from itb to bin.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
SVN-Revision: 45963

9 years agoopenvpn: bump PKG_RELEASE.
Felix Fietkau [Sun, 14 Jun 2015 17:41:54 +0000 (17:41 +0000)]
openvpn: bump PKG_RELEASE.

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
SVN-Revision: 45962

9 years agoopenvpn: let instances drop to nobody in default config.
Felix Fietkau [Sun, 14 Jun 2015 17:41:43 +0000 (17:41 +0000)]
openvpn: let instances drop to nobody in default config.

This is for security precautions.  As persist_tun and persist_key are
already there, this should not cause compatibility issue.

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
SVN-Revision: 45961

9 years agoopenvpn: fix handling option auth_retry.
Felix Fietkau [Sun, 14 Jun 2015 17:41:38 +0000 (17:41 +0000)]
openvpn: fix handling option auth_retry.

As reported in ticket #19104, auth_retry takes a <type> argument with 3
choices: none, nointeract, interact.

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
SVN-Revision: 45960

9 years agolibubox: update to the latest version, adds a few fixes
Felix Fietkau [Sun, 14 Jun 2015 17:41:33 +0000 (17:41 +0000)]
libubox: update to the latest version, adds a few fixes

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 45959

9 years agoar71xx: fix mac offset in firmware file
Felix Fietkau [Sun, 14 Jun 2015 17:41:27 +0000 (17:41 +0000)]
ar71xx: fix mac offset in firmware file

In new firmware-4.bin (for ath10k) mac offset is 0x114 (276), not 0x118 (280).
This also closes #19751, bug introduced in #45624

Signed-off-by: Cezary Jackiewicz <cezary.jackiewicz@gmail.com>
SVN-Revision: 45958

9 years agofirmware-tools/ptgen: use stdint types for struct pte
Felix Fietkau [Sun, 14 Jun 2015 17:41:21 +0000 (17:41 +0000)]
firmware-tools/ptgen: use stdint types for struct pte

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 45957

9 years agofirmware-tools/ptgen: fix endianness conversion for PTE .start, .length field.
Felix Fietkau [Sun, 14 Jun 2015 17:41:16 +0000 (17:41 +0000)]
firmware-tools/ptgen: fix endianness conversion for PTE .start, .length field.

They are of 32-bit little endian integers.

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
SVN-Revision: 45956

9 years agoprocd: correct the order of respawn params in the comments.
Felix Fietkau [Sun, 14 Jun 2015 17:41:07 +0000 (17:41 +0000)]
procd: correct the order of respawn params in the comments.

Signed-off-by: Yousong Zhou <yszhou4tech@gmail.com>
SVN-Revision: 45955

9 years agoar71xx: fix 100/10mbps ethernet link issues on mynet range extender
Felix Fietkau [Sun, 14 Jun 2015 17:41:01 +0000 (17:41 +0000)]
ar71xx: fix 100/10mbps ethernet link issues on mynet range extender

The mynet range extender hardware is suffering from ethernet
link loss when booting with a recent openwrt image. This only
happens on 100mbps links, with 1Gbps speed the link was fine.

The cause of the problem is that the AR8035 PHY (aka F1E)
requires turning on and off the special TX delay setting
depending on the speed of the link.

The 10mbps mode only needed the proper pll value, which was
extracted from the vendor code.

Reported-by: Pascal Paradis
Signed-off-by: Christian Lamparter <chunkeey@googlemail.com>
SVN-Revision: 45954

9 years agosamba36: remove host build paths
Felix Fietkau [Sun, 14 Jun 2015 17:40:52 +0000 (17:40 +0000)]
samba36: remove host build paths

- fix iconv detection because it adds host paths
- disable python detection (host python-config is found)

iconv issue is reported by buildbot config.log + replicated locally
see config.log in logs.tar.gz
python issue observed locally on Arch Linux

Signed-off-by: Dirk Neukirchen <dirkneukirchen@web.de>
SVN-Revision: 45953

9 years agomt76: update to the latest version, adds many fixes
Felix Fietkau [Sat, 13 Jun 2015 08:47:28 +0000 (08:47 +0000)]
mt76: update to the latest version, adds many fixes

Signed-off-by: Felix Fietkau <nbd@openwrt.org>
SVN-Revision: 45952