On Thu, 2014-06-05 at 08:18 +0200, Nikos Mavrogiannopoulos wrote:
Hello,
I'm trying to resubmit the scripts for openconnect in openwrt's luci
interface. Currently the most difficult part in the interface is
specifying the server certificate. There no tools installed by default
in openwrt
Actually, I think we want a kind of 'wizard' for openconnect
configuration in luci. Rather than having hard-coded configuration
items
like 'username' and 'password' which aren't always going to be
relevant,
we actually want to work through the forms that the server offers us.
That's not easily
This fixes a bunch of interoperability issues, and one or two crashes.
Signed-off-by: David Woodhouse dw...@infradead.org
---
net/openconnect/Makefile | 22 +--
.../patches/001-change-openssl-paths.patch | 11 --
2 files changed, 6
they come up — giving me load
balancing across both ADSL lines.
Signed-off-by: David Woodhouse dw...@infradead.org
---
package/iproute2/Makefile |2 ++
package/iproute2/files/30-teql | 23 +++
2 files changed, 25 insertions(+), 0 deletions(-)
create mode 100644
I run this from cron. With remote syslog enabled, this means I have a
record of ADSL line quality variation — which is useful when diagnosing
and reporting faults.
Signed-off-by: David Woodhouse dw...@infradead.org
diff --git a/package/soloscli/Makefile b/package/soloscli/Makefile
index 490186d
On Wed, 2012-03-28 at 16:29 -0700, Dave Taht wrote:
On Wed, Mar 28, 2012 at 3:36 PM, David Woodhouse dw...@infradead.orgwrote:
Resolves https://dev.openwrt.org/ticket/11192
You can now configure devices to be used as a TEQL slave as follows:
# uci set network.ppp0.teql=teql0
# uci set
Resolves https://dev.openwrt.org/ticket/11192
Signed-off-by: David Woodhouse dw...@infradead.org
---
package/iproute2/Makefile |4 +++-
package/iproute2/files/30-teql | 23 +++
2 files changed, 26 insertions(+), 1 deletions(-)
create mode 100644 package/iproute2
Signed-off-by: David Woodhouse dw...@infradead.org
---
.../patches-3.2/120-ppp_txqueue_restart.patch | 115 ++--
.../patches-3.3/120-ppp_txqueue_restart.patch | 115 ++--
2 files changed, 116 insertions(+), 114 deletions(-)
diff --git a/target/linux
Signed-off-by: David Woodhouse dw...@infradead.org
---
target/linux/x86/geos/target.mk |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/target/linux/x86/geos/target.mk b/target/linux/x86/geos/target.mk
index 66c65e0..4da113e 100644
--- a/target/linux/x86/geos/target.mk
Signed-off-by: David Woodhouse dw...@infradead.org
---
.../patches-3.2/130-pppoatm-queue-depth.patch | 182
.../patches-3.3/130-pppoatm-queue-depth.patch | 182
2 files changed, 364 insertions(+), 0 deletions(-)
create mode 100644
target
Some performance improvements here (memcpy elimination) which will be
useful on low-power boxes. And some memory leak fixes too.
Signed-off-by: David Woodhouse dw...@infradead.org
diff --git a/net/openconnect/Makefile b/net/openconnect/Makefile
index a9a6b71..372841c 100644
--- a/net/openconnect
For a start, this isn't really about x86, is it? It's about systems
which use block-based storage instead of flash?
This is for anything which makes the mistake of using CompactFlash or SD
instead of letting the OS have real access to the flash.
There are x86 systems (like OLPCv1, anything with
On Fri, 2012-04-13 at 10:24 +0100, David Woodhouse wrote:
Sounds like you'd be a lot better off using btrfs snapshots for that.
Note that btrfs-progs from the git tree now has a '-r' option to create
an image from a given directly, much like genext2fs does.
--
dwmw2
smime.p7s
Description: S
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
Updated patch with the final commit ID now it's been merged by DaveM.
.../patches-3.2/130-pppoatm-queue-depth.patch | 193
.../patches-3.3/130-pppoatm-queue-depth.patch | 193
2
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
Too slow; my router is running 3.3.2 now...
include/kernel-version.mk |3 +++
target/linux/x86/geos/target.mk |2 +-
2 files changed, 4 insertions(+), 1 deletions(-)
diff --git a/include/kernel-version.mk b/include
We don't allow the targets to be out of sync anyway — for example, if
our 3.3.x kernel is 3.3.2, then no target is allowed to be on 3.3.1.
So don't allow them to set it at all. That way we don't keep ending up
with targets out of sync, when we update and forget to change them all.
If we want to
On Wed, 2012-04-18 at 00:41 -0600, Otto Solares Cabrera wrote:
When was it last updated? Will it ever be updated again? Who decides?
Well, you better take a visit to the bug tracking manager, for a small
un-important feature set (like mine) a kernel upgrade normally works
fine, increase the
On Mon, 2012-04-23 at 02:22 +0200, Michael Markstaller wrote:
Am 23.04.2012 01:52, schrieb Dave Taht:
It should keep working with the --disable-ipv6 switch.
That sounds good ;)
Again: I appreciate your work from a technical point of view, also
understand this might be good at some point in
-script.
It's possible to do it in a hook now, but it also shouldn't be necessary,
because we should be invoking the hotplug scripts to plumb the interface
properly anyway.
Signed-off-by: David Woodhouse dw...@infradead.org
---
net/vpnc-scripts/Makefile | 56
This is very basic — username/password only, and doesn't even check
certificates which it *definitely* should. But it's a good start.
Signed-off-by: David Woodhouse dw...@infradead.org
---
net/openconnect/Makefile |8 ++-
net/openconnect/files/openconnect.sh | 39
On Mon, 2012-04-23 at 10:40 -0700, Bill Moffitt wrote:
Philosophical rant
On the subject of being able to turn off V6, I agree with both sides,
but with a caveat.
On one side, V6 is necessary today in Asia, but, on the other had, it's
not at all necessary or even desirable yet in North
On Mon, 2012-04-23 at 20:14 +0200, Gert Doering wrote:
(Now, what I'm not sure whether OpenWRT already has this: to fully
utilize IPv6 over here, what you need to have is dynamic IPv6 prefix
support using DHCP-PD. As in router queries ISP for a prefix, ISP
assigns 2001:db8:1::/56, router
Signed-off-by: David Woodhouse dw...@infradead.org
---
contrib/package/luci/Makefile |1 +
protocols/openconnect/Makefile |2 +
.../model/cbi/admin_network/proto_openconnect.lua | 62 +++
.../luasrc/model/network
On Mon, 2012-04-23 at 22:32 +0200, Gert Doering wrote:
On Mon, Apr 23, 2012 at 07:43:36PM +0100, David Woodhouse wrote:
It's documented here: http://wiki.openwrt.org/doc/uci/dhcp6c
Oh! This is tremendously cool! Thanks to everyone who made that
possible.
Nit to pick: calling
On Tue, 2012-04-24 at 00:51 +0200, Michael Markstaller wrote:
(I remember the IPv6-day last year, it was funny, even Google failed
30-50%..)
That's an... interesting assertion that I've not heard before. Google's
own analysis was significantly different. They said:
We carried about 65% more
On Tue, 2012-04-24 at 19:13 +1000, Andrew Byrne wrote:
When I created this config file, it was based purely off the existing
WIDE-DHCPv6 config file requirements.
The only way to avoid this variable in the config file will be for
someone to patch the source code so it automatically
On Tue, 2012-04-24 at 07:04 -0400, Brian J. Murrell wrote:
Do you have children? I suspect you don't. Your lack of
forward/future thinking indicates that. It's amazing how one's
perspective of the future changes when one starts thinking about the
future of their children rather than just
On Mon, 2012-04-23 at 07:24 +0200, Felix Fietkau wrote:
Maybe it would make sense to just use the long form
__attribute__((packed)) instead of __packed in parts that are visible to
userspace.
Hm, 'make headers_install' should be performing that substitution
anyway, shouldn't it?
I don't
On Tue, 2012-04-24 at 07:47 -0700, Dave Taht wrote:
b) However __packed leaks out of the main network header file, and was
not defined by iptables to be the define it is in the kernel code.
How? When you run 'make headers_install' to export kernel headers for
use by userspace, it *sanitizes*
On Sun, 2012-04-22 at 14:48 -0700, Dave Täht wrote:
+--- a/net/ipv6/netfilter/ip6t_LOG.c
b/net/ipv6/netfilter/ip6t_LOG.c
+@@ -64,9 +64,9 @@ static void dump_packet(struct sbuff *m,
+ /* Max length: 44 LEN=65535 TC=255 HOPLIMIT=255 FLOWLBL=F */
+ sb_add(m, LEN=%Zu TC=%u
the exported headers.
This should fix it. We do already export the headers for the toolchain,
but I think not the very latest headers, and iptables probably wants
those?
From 99445ea27f5d241ced131b86ae70c2d3c26d943b Mon Sep 17 00:00:00 2001
From: David Woodhouse david.woodho...@intel.com
Date
Each git:// repository is browseable in gitweb by changing git:// to
http://. My collection of patches to openwrt/packages/luci is
accumulating...
===
git://git.infradead.org/users/dwmw2/openwrt.git
David Woodhouse (3):
Add ppp-mod
On Thu, 2012-04-26 at 02:35 +0100, David Woodhouse wrote:
Fix pptp handling of routes to server.
That one probably does want sending as a patch, for review.
I was going to ignore this brokenness, but it ended up actually doing
stupid things on my setup because it was adding a broken
On Fri, 2012-04-27 at 13:23 +0300, Daniel Golle wrote:
Obviously, this goes together with including openl2tp.so plugin which
was omitted in the mini version of the openl2tp package.
I'll have another look into it, maybe it's better to just ommit the ppp setup
timeout entirely in the
On Fri, 2012-04-27 at 15:30 +0300, Daniel Golle wrote:
I wasn't aware of your solution, which I like a lot and which perfectly fits
into how things are done in OpenWrt. Having openl2tp running and handling all
the ppp stuff is pita and just having a plain pppd plugin is a much better
solution
On Sun, 2012-04-22 at 14:48 -0700, Dave Täht wrote:
+
+-#define tcp_flag_word(tp) ( ((union tcp_word_hdr *)(tp))-words [3])
++#define tcp_flag_word2(tp) ( ((union tcp_word_hdr *)(tp))-words [3])
++#define tcp_flag_word(tp) ( __get_unaligned_cpu32union tcp_word_hdr
*)(tp))-words
On Mon, 2012-04-30 at 07:41 -0700, Dave Taht wrote:
Tell it to however wired up this chip and shipped it in qty millions.
Actually that message was already received, successor chipsets from
this manufacturer did it up right.
So the real problem is that the ar71xx doesn't allow you to DMA to
On Mon, 2012-04-30 at 11:11 -0700, Dave Taht wrote:
On Mon, Apr 30, 2012 at 10:26 AM, Felix Fietkau n...@openwrt.org wrote:
On 2012-04-30 5:08 PM, Dave Taht wrote:
On Mon, Apr 30, 2012 at 8:02 AM, Felix Fietkau n...@openwrt.org wrote:
On 2012-04-30 4:49 PM, David Woodhouse wrote:
On Mon
On Mon, 2012-04-30 at 20:34 +0200, Felix Fietkau wrote:
There could be a generic Kconfig variable which could be selected by CPU
targets or ethernet drivers (with a dependency on
!HAVE_EFFICIENT_UNALIGNED_ACCESS).
That's cleaner than messing around with #define stuff manually.
Don't just
On Mon, 2012-04-30 at 19:16 +, Karl P wrote:
As a user, I _never_ want to have to go and turn on the switch that says,
make
my device work as fast as it can
Is there any reason that I would _not_ want this?
You may only have a printer, or something like that, attached to the
crappy
On Mon, 2012-04-30 at 12:20 +0300, Daniel Golle wrote:
I merged your branches into my openwrt, packages and luci repos and
gave it a
try over the past 24 hours running on a ath9k-based router.
Results: very nice. This should go uptream.
Thanks for testing.
I've also retested on the latest
I can give a test account, with IPv6 connectivity through the tunnel, if
anyone needs to be able to test this (for example to make it work with
netif).
Signed-off-by: David Woodhouse dw...@infradead.org
---
contrib/package/luci/Makefile |1 +
protocols/openconnect
Signed-off-by: David Woodhouse dw...@infradead.org
---
.../luasrc/model/cbi/admin_network/proto_l2tp.lua | 57
protocols/ppp/luasrc/model/network/proto_ppp.lua |6 ++-
2 files changed, 62 insertions(+), 1 deletions(-)
create mode 100644 protocols/ppp/luasrc/model
On Fri, 2012-05-04 at 19:55 +0300, Daniel Golle wrote:
PPtP right now works with a pure userland implementation though there is
perfectly working support for kernel-accelerated PPtP.
Having this as a seperate package is a bit ugly, I think it should be a patch
on top of the ppp package adding
On Fri, 2012-05-04 at 23:29 -0600, Philip Prindeville wrote:
I'd combine all of these as:
local device server username password ...
Hm. I was just trying to copy local style, which in ppp.sh had each one
with its usage. Which makes slightly more sense to me; it makes it
simpler to add and
On Fri, 2012-05-04 at 23:35 -0600, Philip Prindeville wrote:
-find_gw() {
- route -n | awk '$1 == 0.0.0.0 { print $2; exit }'
+find_route() {
+ ip route get $1 | sed -e 's/ /\n/g' | \
+sed -ne '1p;/via/{N;p};/dev/{N;p};/src/{N;p};/mtu/{N;p}'
I'd (a) use ip -4 -o
On Sat, 2012-05-05 at 13:14 +0200, Felix Fietkau wrote:
As long as it works, it's not worth putting in effort to clean this up.
This code will be going away soon (replaced by the to-be-written netifd
based script).
The new code will not need iproute2 or such tricks at all.
OTOH vpnc-script
On Sat, 2012-05-12 at 14:17 -0600, Philip Prindeville wrote:
Gabor is retiring old versions of 2.6.x regularly (thanks, Gabor!).
Alas, he's one step away from retiring 2.6.39.4 which is what x86 is
currently at.
Does anyone else find it troubling that x86 has been as stagnant as it is?
On Thu, 2012-05-24 at 13:45 +0100, Conor O'Gorman wrote:
The default a2p (adsl2+) is perhaps too strict? The 'a' option
includes all 'a' variations:
annex_a=04_01_04_00_00_01_00_00
annex_at1=01_00_00_00_00_00_00_00
annex_alite=00_01_00_00_00_00_00_00
annex_admt=04_00_00_00_00_00_00_00
On Fri, 2012-11-02 at 01:47 -0700, Russell Senior wrote:
Nathan == Nathan Williams nat...@traverse.com.au writes:
Nathan Hi, I've just noticed that since r33640 (x86: remove
Nathan grub-legacy) my builds for Geos don't boot. After the GRUB
Nathan 2.00 menu on the serial port, it only gets
On Wed, 2012-11-28 at 14:35 -0800, Daniel Gimpelevich wrote:
The br2684 driver seems to expect that the MAC is already on the device
before a virtual circuit is created, but it seems to leave the mechanism
for doing so to the ATM driver. The only other working ATM driver in
OpenWrt is for AR7,
-by: David Woodhouse david.woodho...@intel.com
---
For AA too.
Config.in|7 +++
include/image.mk |2 +-
2 files changed, 8 insertions(+), 1 deletions(-)
diff --git a/Config.in b/Config.in
index 46f7ce8..76acbe5 100644
--- a/Config.in
+++ b/Config.in
@@ -125,6 +125,13 @@ menu
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
target/linux/generic/config-3.6 |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/target/linux/generic/config-3.6 b/target/linux/generic/config-3.6
index 5af13b6..7af4f6e 100644
--- a/target/linux/generic/config
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
How do I set KERNEL_VERSION:=3.6.7 just for Geos? In geos/target.mk it doesn't
seem to work...
target/linux/x86/config-3.6 | 386 ++
target/linux/x86/geos/config-3.6 | 25 +++
2 files
On Wed, 2012-11-28 at 16:28 -0800, Daniel Gimpelevich wrote:
On Thu, 2012-11-29 at 00:17 +, David Woodhouse wrote:
I mostly use that with PPPoA, rather than PPPoEoA with the pointless MTU
breakage that that implies. But it does work fine with BR2684 with the
MAC address 00:00:01:00:00
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
On Thu, 2012-11-29 at 00:25 +, David Woodhouse wrote:
How do I set KERNEL_VERSION:=3.6.7 just for Geos? In geos/target.mk it
doesn't seem to work...
Yes it does; I just need to make target/kernel/clean because the
dependencies
Patches queued upstream in net-next.git
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
The 3.3 patch for AA too.
.../generic/patches-3.3/065-8139cp-fixes.patch | 189
.../generic/patches-3.6/065-8139cp-fixes.patch | 176
On Thu, 2012-11-29 at 18:47 +0100, Gabor Juhos wrote:
2012.11.29. 1:23 keltezéssel, David Woodhouse írta:
Patches about to go into net-next.git
Signed-off-by: David Woodhouse david.woodho...@intel.com
Applied this one, however it seems that 3/5 is lost somewhere.
Um... my own outbound
On Thu, 2012-11-29 at 20:33 +0100, Gabor Juhos wrote:
Should we go ahead and change the default for
CONFIG_TARGET_ROOTFS_RESERVED_PCT to 0? I can't see any good reason for
leaving it at 5% (except for the initial patch which just added the option
without changing anything, on
On Thu, 2012-11-29 at 15:17 -0500, John Lauro wrote:
I tried preinit and also
.../trunk/package/base-files/files/etc/config/network
.../trunk/target/linux/ar7/base-files/etc/config/network
and it still comes out as 192.168.1.1. (building for a wndr3800)
Seems to be lots of network
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
Config.in |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/Config.in b/Config.in
index 8a00412..ecbfb04 100644
--- a/Config.in
+++ b/Config.in
@@ -128,7 +128,7 @@ menu Target Images
config
* of the boot partition, which
is entirely pointless when both it and the rootfs are ext4. So I had a
go at fixing that...
Does this seem reasonable?
From 458d58b0ca667a28085657416fb1f0b9dedcb85a Mon Sep 17 00:00:00 2001
From: David Woodhouse david.woodho...@intel.com
Date: Mon, 3 Dec 2012 01:12:32
On Thu, 2012-12-06 at 04:42 -0500, Weedy wrote:
Anyone flashed recent trunk builds? It's refusing to take my factory.img.
Tomorrow I'll try a backfire image or something.
Not trunk, but I put AA (r34386) on a new WNDR3800 last week. The stock
firmware took it just fine.
(I then thought I'd
On Mon, 2012-08-13 at 15:06 +0200, Florian Fainelli wrote:
On Thursday 26 April 2012 16:44:02 Abaakouk Mehdi wrote:
RESENT: Updated patch with last remarks fixed
Makefile and patch for package NUT 2.6.3
Signed-off-by: Mehdi Abaakouk sil...@sileht.net
I do not see any drivers in
From: Mehdi Abaakouk sil...@sileht.net
Makefile and patch for package NUT 2.6.5
dwmw2: Update to NUT 2.6.5, add OpenUPS, tidy up per-driver build a little.
Signed-off-by: Mehdi Abaakouk sil...@sileht.net
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
utils/nut/Makefile
On Fri, 2012-12-07 at 12:34 -0800, Daniel Gimpelevich wrote:
Index: target/linux/lantiq/base-files/etc/init.d/ifx-esi
===
--- target/linux/lantiq/base-files/etc/init.d/ifx-esi (revision 0)
+++
On Fri, 2012-12-07 at 12:59 -0800, Daniel Gimpelevich wrote:
It'd be better to put that in /etc/hotplug.d/atm so that you don't have
to worry about the timing of when the atm driver loads at startup, and
it gets done properly if the driver is ever unloaded and reloaded, etc.
OK, but
On Fri, 2012-12-07 at 18:58 -0800, Daniel Gimpelevich wrote:
OK, but hotplug never generates any atm events, so Felix suggested this
instead.
Hm, it should do. We have other scripts in /etc/hotplug.d/atm, and
hotplug2.rules definitely *looks* like it should be invoking them.
On Thu, 2012-12-06 at 21:12 +, David Woodhouse wrote:
It could do with a little more work, ideally. We should have an
initscript which generates the config files (like ups.conf) in tmpfs
automatically from uci config.
How does this look?
From 4295de79d71e4f5a317b37c23cb261c61e23 Mon
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
utils/nut/Makefile | 11 ++-
1 files changed, 10 insertions(+), 1 deletions(-)
diff --git a/utils/nut/Makefile b/utils/nut/Makefile
index 9d9ffd1..3fcfb7e 100644
--- a/utils/nut/Makefile
+++ b/utils/nut/Makefile
@@ -75,6
This works now...
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
utils/collectd/Makefile |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/utils/collectd/Makefile b/utils/collectd/Makefile
index 413ff84..6e609b0 100644
--- a/utils/collectd/Makefile
+++ b
--- /dev/null
+++ b/applications/luci-statistics/luasrc/model/cbi/luci_statistics/nut.lua
@@ -0,0 +1,28 @@
+--[[
+LuCI - Lua Configuration Interface
+
+Copyright © 2011 Manuel Munz freifunk at somakoma dot de
+Copyright © 2012 David Woodhouse dw...@infradead.org
+
+Licensed under the Apache License
On Mon, 2012-12-10 at 17:26 +, David Woodhouse wrote:
+ -- Note: This is in ISO8859-1 for rrdtool. Welcome to the 20th
century.
+ local temperature = {
+ title = %H: Battery temperature on UPS \%pi\,
+ vlabel = °C,
+ number_format
On Sat, 2012-12-08 at 18:08 +0100, Florian Fainelli wrote:
On Thursday 06 December 2012 21:17:01 David Woodhouse wrote:
From: Mehdi Abaakouk sil...@sileht.net
Makefile and patch for package NUT 2.6.5
dwmw2: Update to NUT 2.6.5, add OpenUPS, tidy up per-driver build a
little
at somakoma dot de
+Copyright © 2012 David Woodhouse dw...@infradead.org
+
+Licensed under the Apache License, Version 2.0 (the License);
+you may not use this file except in compliance with the License.
+You may obtain a copy of the License at
+
+ http://www.apache.org/licenses/LICENSE-2.0
This makes it consistent with the apcups and snmp plugins. And makes the
max value of 3600 slightly less insane (2½ days, not 1 hour).
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
utils/collectd/Makefile|2 +-
.../patches/130-nut-timeleft
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
.../generic/patches-3.6/065-8139cp-fixes.patch | 49 ---
1 files changed, 41 insertions(+), 8 deletions(-)
diff --git a/target/linux/generic/patches-3.6/065-8139cp-fixes.patch
b/target/linux/generic/patches-3.6
Same patch for 3.3, 3.6 and 3.7. And now my shiny new DMA-capable Geos
no longer crashes with a string of oops at boot...
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
For AA too.
.../linux/generic/patches-3.3/132-solos-dma.patch | 37
.../linux/generic
This works now...
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
utils/collectd/Makefile |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/utils/collectd/Makefile b/utils/collectd/Makefile
index 413ff84..6e609b0 100644
--- a/utils/collectd/Makefile
+++ b
There's more to be done here, but this is a good start and covers most
things that people will need to configure.
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
utils/nut/Makefile | 16 +++---
utils/nut/files/ups.init| 87 ++
utils/nut/files
From: David Woodhouse david.woodho...@intel.com
This almost caused a nasty accident on my Geos — without the MSR access
it managed to erase *most* of the flash and write the new image, but
didn't manage to change the last few KiB. Thankfully I'd already taken a
copy of the old contents, and could
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
utils/nut/Makefile | 11 ++-
1 files changed, 10 insertions(+), 1 deletions(-)
diff --git a/utils/nut/Makefile b/utils/nut/Makefile
index 9d9ffd1..3fcfb7e 100644
--- a/utils/nut/Makefile
+++ b/utils/nut/Makefile
@@ -75,6
From: David Woodhouse david.woodho...@intel.com
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
.../linux/generic/patches-3.3/132-solos-dma.patch | 38
.../linux/generic/patches-3.6/132-solos-dma.patch | 38
.../linux/generic/patches
On Wed, 2012-12-12 at 21:56 +, David Woodhouse wrote:
From: David Woodhouse david.woodho...@intel.com
This almost caused a nasty accident on my Geos — without the MSR access
it managed to erase *most* of the flash and write the new image, but
didn't manage to change the last few KiB
On Wed, 2012-12-12 at 21:56 +, David Woodhouse wrote:
This makes it consistent with the apcups and snmp plugins. And makes the
max value of 3600 slightly less insane (2½ days, not 1 hour).
Upstream have done this differently, by making everything use *seconds*
not minutes. This means
plugins, those would probably want fixing to multiply by 60.
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
utils/collectd/Makefile |2 +-
utils/collectd/patches/130-timeleft-max.patch | 14 ++
2 files changed, 15 insertions(+), 1 deletions
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
utils/nut/Makefile | 11 ++-
1 files changed, 10 insertions(+), 1 deletions(-)
diff --git a/utils/nut/Makefile b/utils/nut/Makefile
index 9d9ffd1..3fcfb7e 100644
--- a/utils/nut/Makefile
+++ b/utils/nut/Makefile
@@ -75,6
There's more to be done here, but this is a good start and covers most
things that people will need to configure.
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
utils/nut/Makefile | 16 +++---
utils/nut/files/ups.init| 87 ++
utils/nut/files
This almost caused a nasty accident on my Geos — without the MSR access
it managed to erase *most* of the flash and write the new image, but
didn't manage to change the last few KiB. Thankfully I'd already taken a
copy of the old contents, and could flash them back.
Signed-off-by: David Woodhouse
On Sat, 2012-12-15 at 11:56 +0100, John Crispin wrote:
i pushed the lantiq 3.7 kernel support last night.
A bunch of things break with the 3.7 kernel. This addresses a few of
them, but there's more. In particular, a lot of NAT stuff is no longer
specific to Legacy IP — we have IPV6 NAT support
On Sun, 2012-12-16 at 11:47 +0100, John Crispin wrote:
On 16/12/12 11:21, David Woodhouse wrote:
On Sat, 2012-12-15 at 11:56 +0100, John Crispin wrote:
i pushed the lantiq 3.7 kernel support last night.
A bunch of things break with the 3.7 kernel. This addresses a few of
them
From: David Woodhouse david.woodho...@intel.com
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
package/network/utils/ipset/Makefile |4 ++--
1 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/package/network/utils/ipset/Makefile
b/package/network/utils/ipset
From: David Woodhouse david.woodho...@intel.com
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
.../604-netfilter_cisco_794x_iphone.patch | 22 ---
1 files changed, 14 insertions(+), 8 deletions(-)
diff --git
a/target/linux/generic/patches-3.7/604
From: David Woodhouse david.woodho...@intel.com
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
target/linux/x86/geos/target.mk |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/target/linux/x86/geos/target.mk b/target/linux/x86/geos/target.mk
index 4afb34a
From: David Woodhouse david.woodho...@intel.com
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
include/netfilter.mk | 17 +++--
1 files changed, 11 insertions(+), 6 deletions(-)
diff --git a/include/netfilter.mk b/include/netfilter.mk
index e166628..20bfffe 100644
From: David Woodhouse david.woodho...@intel.com
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
package/kernel/modules/crypto.mk |7 +--
1 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/package/kernel/modules/crypto.mk b/package/kernel/modules/crypto.mk
index
From: David Woodhouse david.woodho...@intel.com
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
target/linux/x86/config-3.7| 389 ++
.../linux/x86/patches-3.7/006-yenta_mistery.patch | 20 +
.../patches-3.7/009-rdc321x_select_embedded.patch
From: David Woodhouse david.woodho...@intel.com
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
target/linux/generic/config-3.7 |1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/target/linux/generic/config-3.7 b/target/linux/generic/config-3.7
index b572515
Signed-off-by: David Woodhouse david.woodho...@intel.com
---
Oops. This was definitely here earlier, but I evidently forgot to commit it.
target/linux/x86/geos/config-3.7 | 25 +
1 files changed, 25 insertions(+), 0 deletions(-)
create mode 100644 target/linux/x86/geos
a bit off a bunch of modules too. No size win on x86 — the
movbe instruction takes as many bytes to encode as the equivalent bswap
+ mov. But it's allegedly a significant *performance* win on Atom, which
is why it was added there in the first place.
Signed-off-by: David Woodhouse david.woodho
1 - 100 of 179 matches
Mail list logo
