Re: [OpenWrt-Devel] printk() and browser
I have figure the problem, but I don't know the solution.
Please if someone can tell me about why the following errors are given on
dmesg.
iptable_raw:Unknow symbol xt_hook_link (err 0)
iptable_raw:Unknow symbol ipt_alloc_initial_table (err 0)
iptable_raw:Unknow symbol xt_hook_unlink (err 0)
xt_NOTRACK: Unknow symbol xt_register_target (err 0)
xtconntrack: Unknown symbol xt_register_match(err0)
xtconntrack: Unknown symbol xt_unregister_match(err0)
There are a lot of such errors on dmesg,
when i compile the following lines of code in at_rx_tasklet() in recv.c
static int ai=0;
if (ai==0) {
printk("abhinav\n"); ai++ ;}
I think i have to do some macro unset or something else ?
Can anyone please help...
whoever is trying to patch the kernel
Abhinav
On Mon, Oct 17, 2011 at 6:38 PM, abhinav narain
wrote:
> hi,
> I am modifying the driver for past few days and noticed something today.
> Though my modifications are seen on the command line, I have one thing
> going extremely worn i realized today.
> The vanilla installation of the firmware on the router is perfect.
> I am able to browse the internet, but when I even do a printk() in the
> tasklet function (just once by using a condition),
> my laptop is unable to connect to the Internet !
>
> What can be going wrong ?
> /etc/resolv.conf is the same for both the cases :
> search lan
> nameserver 127.0.0.1
>
> Please help.
>
> Abhinav
>
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
[OpenWrt-Devel] combined extended image
Hi folks,
being a regular user of the combined image I'd like to propose changing the
format in order to render it more flexible. The main focus are the embedded
files: The new format shall be able to embedded numerous files and their
filenames. Furthermore, the type field shall make the image identifiable even
if the image name has been mangled.
Here is what I had in mind:
# Write image header followed by all specified files
# The header is padded to 64k, format is:
# CE magic word ("Combined Extended Image") (2 bytes)
# short description of the target device (10 bytes)
#number of files following the header (2 byte)
# name of the first file (20 bytes)
# length of the first file encoded as zero padded 8 digit
hex (8 bytes)
# name of the Nth file (20 bytes)
# length of the Nth file encoded as zero padded 8 digit hex
(8 bytes)
Attached you will find a script to generate the format I described.
Cheers,
Marek
combined-ext-image.sh
Description: application/shellscript
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
[OpenWrt-Devel] printk() and browser
hi, I am modifying the driver for past few days and noticed something today. Though my modifications are seen on the command line, I have one thing going extremely worn i realized today. The vanilla installation of the firmware on the router is perfect. I am able to browse the internet, but when I even do a printk() in the tasklet function (just once by using a condition), my laptop is unable to connect to the Internet ! What can be going wrong ? /etc/resolv.conf is the same for both the cases : search lan nameserver 127.0.0.1 Please help. Abhinav ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] releasing the PBX interface: luci-app-pbx
On Mon, Oct 17, 2011 at 3:40 PM, i iordanov wrote: > The same link to the code was updated to also deal with dynamic IP addresses. For the record, it was: http://dancho.no-ip.com/pbx/luci-app-pbx-0.11.tar.gz Iordan -- The conscious mind has only one thread of execution. ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] releasing the PBX interface: luci-app-pbx
Hi Jow, On Mon, Oct 17, 2011 at 1:33 PM, Jo-Philipp Wich wrote: > Yes, thats probably enough. The same link to the code was updated to also deal with dynamic IP addresses. Cheers, Iordan -- The conscious mind has only one thread of execution. ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
[OpenWrt-Devel] [PATCH] openl2tp - finally enable config file option
Until now the openl2tpg package only allowed configuration through RPC
by using the l2tpconfig command, making many some tasks difficult.
Discussion exists at https://dev.openwrt.org/ticket/10164 . Enabling the
config file option requires working around the flex library ( (libfl)
and a few compile fixes. The following 2 patches enable the config file
option in addition to the RPC option. Tested against trunk r28428. Note
that the openwrt openl2tp package is still not fully finished, e.g. a
proper init.d script and sample conf files should be included, see
build_dir/*/openl2tp-1.8/{etc,ipsec}/*. I may sort that out as well later.
Signed-off-by: Bogdan Roman
File: 020-makefile-conf-file.patch
--- a/Makefile
+++ b/Makefile
@@ -8,7 +8,7 @@ L2TP_FEATURE_LAIC_SUPPORT= y
L2TP_FEATURE_LAOC_SUPPORT= y
L2TP_FEATURE_LNIC_SUPPORT= y
L2TP_FEATURE_LNOC_SUPPORT= y
-#L2TP_FEATURE_LOCAL_CONF_FILE= y
+L2TP_FEATURE_LOCAL_CONF_FILE= y
#L2TP_FEATURE_LOCAL_STAT_FILE= y
# Define USE_DMALLOC to enable dmalloc memory debugging
@@ -204,7 +204,7 @@ ifeq ($(L2TP_FEATURE_LOCAL_CONF_FILE),y)
# Config file parser
LEX= flex
YACC= bison
-LDFLAGS.l2tpd+=-lfl
+#LDFLAGS.l2tpd+= -lfl
%.c: %.l
$(LEX) -o$@ $<
@@ -212,10 +212,10 @@ LDFLAGS.l2tpd+= -lfl
$(YACC) -d -o l2tp_config_parse.c $<
l2tp_config_token.o: l2tp_config_token.c
- $(CC) -I. -MMD -w $(CFLAGS.optimize) -c -DYY_NO_UNPUT $<
+ $(CC) -I. -MMD -w $(CFLAGS.optimize) -c $(CPPFLAGS)
-DYY_NO_UNPUT $<
l2tp_config_parse.o: l2tp_config_parse.c l2tp_config_parse.h
- $(CC) -I. -MMD -w $(CFLAGS.optimize) -c -DYY_NO_UNPUT $<
+ $(CC) -I. -MMD -w $(CFLAGS.optimize) -c $(CPPFLAGS)
-DYY_NO_UNPUT $<
endif
# Compile without -Wall because rpcgen-generated code is full of warnings.
File: 020-conf-no-libfl.patch
--- a/l2tp_config_token.l
+++ b/l2tp_config_token.l
@@ -23,6 +23,8 @@ static int resolveip(const char *val, st
int lineno;
int command;
.
+int yywrap(void) { return 1; }
+
%}
.
/* common section */
--
Bogdan Roman, http://www.cl.cam.ac.uk/~abr28/
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] releasing the PBX interface: luci-app-pbx
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Yes, thats probably enough. ~ Jow -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk6cZu8ACgkQdputYINPTPN65QCdGD/8SjP4H5BiIlww5Cd3hUZr t6IAn3hmizfBu1h3TJJuAsU6h7tRTyUr =ArGh -END PGP SIGNATURE- ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] releasing the PBX interface: luci-app-pbx
Hi Jow,
On Mon, Oct 17, 2011 at 12:22 PM, Jo-Philipp Wich wrote:
> you'll get the IP address behaviour if "lan" is set to proto dhcp or
> something else which does not specify an ipaddr in advance.
This makes sense. However, it also makes sense to not support dynamic
IP addresses, because if it changes, the configuration of all clients
will be invalid. If the user is advanced enough to have local DNS, and
to put in a hostname in the client configuration, then said user would
not need the output of this field anyway.
My thoughts are that if I get an empty string after quering
("network", "lan", "ipaddr"), I can print a short sentence that reads
something along these lines:
"IP address not static. Please set a static IP address in Network->LAN."
or even simply "IP address not static."
Would this work any better than just leaving the field empty?
Cheers,
Iordan
--
The conscious mind has only one thread of execution.
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] releasing the PBX interface: luci-app-pbx
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Hi, you'll get the IP address behaviour if "lan" is set to proto dhcp or something else which does not specify an ipaddr in advance. ~ Jow -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk6cVmEACgkQdputYINPTPM70QCeInqJGV3FRz4wGT5D7I6yJ8W9 cL4AnjPZVrNC4eHwUZ86AvOm7QyTk6zU =MmD9 -END PGP SIGNATURE- ___ openwrt-devel mailing list openwrt-devel@lists.openwrt.org https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] [PATCH] hiawatha: update to 7.7
Applied in [28479], thanks !
--
-{Nico}
2011/10/13 Raphaël HUCK :
> Hi all,
>
> this patch updates hiawatha to 7.7. Changelog:
>
> * First parameter of Alias can now contain subdirectories.
> * Improved stability for connections with SSL client authentication.
> * Bugfix: BanOnFlooding was broken.
>
> Thanks!
>
> -Raphaël
>
> ___
> openwrt-devel mailing list
> openwrt-devel@lists.openwrt.org
> https://lists.openwrt.org/mailman/listinfo/openwrt-devel
>
>
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
Re: [OpenWrt-Devel] releasing the PBX interface: luci-app-pbx
Hi Jow,
I had a bit of time over the weekend to try some stuff. I've changed
the behavior of pbx-calls.lua to now display a string saying what the
system will do instead of leaving the box empty (i.e. "Dials all
numbers"). The only drawback is that one would either have to delete
the string when adding prefixes, or hit enter and add the prefixes
after, and the the app will throw out the "Dials all numbers" string
since it is an "invalid" prefix.
I have not changed anything to do with ipaddr, since I got no
indication from anyone how I can reproduce this, or that my code is
wrong to get ipaddr from ("network", "lan", "ipaddr").
Here is a link to the new code:
http://dancho.no-ip.com/pbx/luci-app-pbx-0.11.tar.gz
Thanks!
Iordan
--
The conscious mind has only one thread of execution.
___
openwrt-devel mailing list
openwrt-devel@lists.openwrt.org
https://lists.openwrt.org/mailman/listinfo/openwrt-devel
