[OpenWrt-Devel] WPA-NONE to secure IBSS Ad-Hoc networks

2010-10-05 Thread Daniel Golle
Hi! I'm trying to setup an Ad-Hoc networking using one common shared AES-256 (CCMP) key. Is WPA-NONE supported in OpenWrt? What's the intended way of securing an Ad-Hoc IBSS network with a shared key? In the Forum, the question remains unanswered https://forum.openwrt.org/viewtopic.php?id=1607

[OpenWrt-Devel] Lua number representation: request for rollback

2010-10-05 Thread Rooslan S. Khayrov
Hi all, In r18159 representation of floating point numbers in the LNUM patch was changed from double to float. This has side effects which I find very hard to justify. 32-bit integer cannot be represented as a single-precision floating point number without loss of precision. We've run into this p

Re: [OpenWrt-Devel] Infineon Danube/TwinPass with open DSL & VoIP drivers (ifxmips)

2010-10-05 Thread John Crispin
i am going to spent a few days next week working on the danube support, so i will do the arv4518 at the same time. i will also be merging ata support whilst at it :D thx for finding the patches Luca Olivetti wrote: > En/na Luca Olivetti ha escrit: >> Al 30/09/10 19:53, En/na John Crispin ha

Re: [OpenWrt-Devel] Infineon Danube/TwinPass with open DSL & VoIP drivers (ifxmips)

2010-10-05 Thread Luca Olivetti
En/na Luca Olivetti ha escrit: Al 30/09/10 19:53, En/na John Crispin ha escrit: we can make uboot for many of the other arcaydian boards and i will be adding them to the wiki next week. however for the ARV4518 and ARV4519 we have no uboot yet due to a missing switch driver. i hope to find the

Re: [OpenWrt-Devel] [PATCH] ramips: Allow ethernet interface to be taken down and up again

2010-10-05 Thread Gabor Juhos
2010.10.05. 12:15 keltezéssel, Helmut Schaa írta: > Taking the ramips ethernet interface down and up again resulted in the > driver not receiving any frames anymore. Fix this by correctly disabling > interrupts in the hw on ifdown. > > Signed-off-by: Helmut Schaa Applied in r23243. Thanks, Gabo

Re: [OpenWrt-Devel] Is there a way to verify an image before flashing it to the route (and bricking it)?

2010-10-05 Thread Alexey Loukianov
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 13.09.2010 19:40, Esa Hyytia wrote: > ... to compile the firmware with "ramdisk" option, upload the image to > RAM from bootloader (via serial interface) using tftp, and booting the > image from RAM without flashing it at all. This works nicely for me

Re: [OpenWrt-Devel] Fwd: OpenWRT build system: how to run configure for a single package?

2010-10-05 Thread Ralph.Hempel
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 > Looks like this is a bug in the OpenWRT build system, it might be > reasonable to > fire the bug report against this. ... it's just an enhancement for source ball developers (like dist & distcheck) . Let me add it today night to svn . br Ralph -

Re: [OpenWrt-Devel] Fwd: OpenWRT build system: how to run configure for a single package?

2010-10-05 Thread Alexey Loukianov
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 05.10.2010 15:54, ralph.hem...@lantiq.com wrote: > core$hg diff include/subdir.mk > diff --git a/include/subdir.mk b/include/subdir.mk > --- a/include/subdir.mk > +++ b/include/subdir.mk > @@ -5,7 +5,7 @@ > # See /LICENSE for more information. > # >

Re: [OpenWrt-Devel] Fwd: OpenWRT build system: how to run configure for a single package?

2010-10-05 Thread Ralph.Hempel
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 > would allow me to catch build configuration mistakes earlier and don't > waste a > lot of time on debugging later. core$hg diff include/subdir.mk diff --git a/include/subdir.mk b/include/subdir.mk - --- a/include/subdir.mk +++ b/include/subdir.mk @

[OpenWrt-Devel] [PATCH] ramips: Allow ethernet interface to be taken down and up again

2010-10-05 Thread Helmut Schaa
Taking the ramips ethernet interface down and up again resulted in the driver not receiving any frames anymore. Fix this by correctly disabling interrupts in the hw on ifdown. Signed-off-by: Helmut Schaa --- target/linux/ramips/files/drivers/net/ramips.c |4 1 files changed, 4 insertion

Re: [OpenWrt-Devel] [PATCH] Define recent gcc for x86 target

2010-10-05 Thread Imre Kaloz
I'll bump the compilers for way more targets after some testing, then nuke 4.1.2. Imre On Mon, 04 Oct 2010 21:27:47 +0200, Roberto Riggio wrote: Current Config.version does not define a gcc version for the x86 target. As a result version 4.1.2 is used. Signed-off-by: Roberto Riggio --

Re: [OpenWrt-Devel] problem with /usr/lib/opkg/status

2010-10-05 Thread Jo-Philipp Wich
Hi. I pushed your opkg patch upstream, see http://code.google.com/p/opkg/source/detail?r=576. Opkg was updated in trunk at r23230, the sysupgrade integration was added in r23233. Both changes have been merged to the Backfire branch. I still left the whole /etc/config/ in place since some file

Re: [OpenWrt-Devel] Fwd: OpenWRT build system: how to run configure for a single package?

2010-10-05 Thread Alexey Loukianov
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 05.10.2010 10:42, Alexey Loukianov wrote: > What about doing the following instead: > > make package/helloworld/{clean,prepare,compile,install} V=99 ? > >> >> Am I missing something? > > The 'configure' target you are looking for is 'prepare' > Un

[OpenWrt-Devel] [PATCH] Remove fhonour-copts

2010-10-05 Thread Roberto Riggio
As suggested by Florian, I'm submitting a patch that removes the honour-copts CFLAG when using an external toolchain. With the flag I'm getting this error when trying to compile a package (the click router): cc1plus: error: unrecognized command line option "-fhonour-copts" Signed-off-by: Rober

Re: [OpenWrt-Devel] Disable honour-copts flag

2010-10-05 Thread Roberto Riggio
On 10/04/2010 11:39 PM, Florian Fainelli wrote: Hi Roberto, Hi I suppose you are using an external toolchain right? This is something that confuses me. I see from the package-default file that the flag honour-copts is specified only when external toolchain is used. However, I'm using the defa