Re: [OpenWrt-Devel] [PATCH] ar71xx: fix WNDR3x00 factory image generation

2011-12-19 Thread Jonathan McCrohan
On 18/12/11 23:14, Olipro wrote: > The WNDR3700 should be split entirely from the WNDR3700v2/3800 - having > them in the same profile will cause a build failure if you have configured > enough packages to push you over the 8MB limit of the 3700 - I have > currently modified my scripts so that al

[OpenWrt-Devel] Bluez 4

2011-12-19 Thread MailingList Aggregator
Hi, is there a way to build latest Bluez (bluez-4.96.tar.gz) instead of (bluez-libs-3.36.tar.gz + bluez-utils-3.36.tar.gz) ? Currently I've successfully built the bluetooth system, using make menuconfig and make kernel_menuconfig. I have modules and tools. I can detect devices with # hciconfig -a

Re: [OpenWrt-Devel] [PATCH 1/1] backport of -o local_lock option for nfs mounts

2011-12-19 Thread Pete Holland
No worries. I have been sitting on this (and the other two I just submitted) to make sure they were working in our production environment. I should have submitted them earlier. At least they are here in case another dev runs into the same problems and needs to fix them. The NFS hang is particula

[OpenWrt-Devel] [PATCH 1/1] fix connection tracking helper logging on invalid frames

2011-12-19 Thread Pete Holland
This patch fixes a possible DoS where invalid frames that use conntrack helpers can continuously log to the kernel log. This could be a particular problem for embedded systems with small or slow filesystems. Fix is to use the LOG_INVALID macro which only logs if the proper sysctl setting in enable

[OpenWrt-Devel] [PATCH 1/1] Backport SUNRPC NFS hang fixes

2011-12-19 Thread Pete Holland
The following is a backport of SUNRPC fix for a NFS hang (Kernel Bug 16494) and related fixes for SYN_SENT state handling and proper clearing of the srcport field in the SUNRPC transport. Signed-off-by: Peter Holland --- --- /dev/null Thu Jan 01 00:00:00 1970 + +++ b/target/linux/generic-

Re: [OpenWrt-Devel] [PATCH 1/1] backport of -o local_lock option for nfs mounts

2011-12-19 Thread Jo-Philipp Wich
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, thanks for you patch, unfortunately I cannot accept kernel level patches for Backfire anymore. If we do another point release after 10.03.1 we'll look into it. ~ Jow -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using

[OpenWrt-Devel] [PATCH 1/1] backport of -o local_lock option for nfs mounts

2011-12-19 Thread Pete Holland
This patch is a backport of the -o local_lock option for nfs mounts. The local_lock option was introduced into the mainline in kernel 2.6.37. Use of -o local_lock=flock is required to avoid deadlocks when running SMB over NFS. >From the updated nfs(5) manpage: To support legacy flock behavior sim

Re: [OpenWrt-Devel] [PATCH] Add enabled option to multiwan

2011-12-19 Thread Etienne Champetier
Hi Here follows the patch for luci multiwan package I've also opened a ticket on the luci trac Regards Le 19/12/2011 16:38, Etienne Champetier a écrit : > Hi > > The only mean to disable multiwan is to delete the symlink in /etc/rc.d, but > if you make a backup and restore it you have to re-del

Re: [OpenWrt-Devel] PATCH: Buffalo WZR-HP-G300NH2 support

2011-12-19 Thread Peter Naulls
On 12/19/2011 01:04 AM, Felix Fietkau wrote: On 2011-12-19 6:31 AM, Mark Deneen wrote: On Mon, Dec 19, 2011 at 12:25 AM, Mark Deneen wrote: On Sun, Dec 18, 2011 at 5:20 PM, Felix Fietkau wrote: Applied the patch with my modifications in r29567 - Felix Will this be built along with the oth

[OpenWrt-Devel] [PATCH] Add enabled option to multiwan

2011-12-19 Thread Etienne Champetier
Hi The only mean to disable multiwan is to delete the symlink in /etc/rc.d, but if you make a backup and restore it you have to re-delete the symlink. This small patch add an "enabled" option to multiwan script. It will allow me to have one image per router model, not one with multiwan and one

Re: [OpenWrt-Devel] RFC implementing wifi client connection management

2011-12-19 Thread Christian Kapeller
> So rework mac80211/wpa_supplicant integration to work like hostapd, > there we also just define a base part and append additional sections for > each wifi-iface. I will do that. Does this feature, implemented right, have a chance to go upstream? If so, I'll keep in touch about the implementation

Re: [OpenWrt-Devel] RSSI LEDs

2011-12-19 Thread Daniel Golle
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 12/19/2011 11:10 AM, Felix Fietkau wrote: > I'm actually considering handling RSSI LED in user space. It doesn't > integrate well with the normal trigger <-> LED model, handling it It's not that far from what ledtrig-usbdev does in the end. Just tha

Re: [OpenWrt-Devel] RSSI LEDs

2011-12-19 Thread Felix Fietkau
On 2011-12-19 8:17 AM, Daniel Golle wrote: > Hi! Various devices got signal strength indicator LEDs connected to > the SoCs GPIO PINs. Obviously it's easy to control them with a > polling process using libiwinfo, but that's far from a perfect > solution (it's ugly, it's an additional process, it's

Re: [OpenWrt-Devel] PATCH: Buffalo WZR-HP-G300NH2 support

2011-12-19 Thread Felix Fietkau
On 2011-12-19 6:31 AM, Mark Deneen wrote: > On Mon, Dec 19, 2011 at 12:25 AM, Mark Deneen wrote: >> On Sun, Dec 18, 2011 at 5:20 PM, Felix Fietkau wrote: >>> Applied the patch with my modifications in r29567 >>> >>> - Felix >> >> Will this be built along with the other nightly builds? >> >> -M >