gmane.linux.drivers.gnokii

Re: Cross Compiling Gnokii for ARM

Daniele Forsi — 2012-05-08T15:48:21

2012/5/8 Forconi:


shouldn't this also contain -lglib-2.0 ?

did you try to make pkg-config find its configuration with this
environment variable instead?
"""
PKG_CONFIG_SYSROOT_DIR
              Modify  -I and -L to use the directories located in
target sysroot.  this option is useful when crosscompiling package
that use pkg-config to determine CFLAGS
              anf LDFLAGS. -I and -L are modified to point to the new
system root. this means that a -I/usr/include/libfoo will become
-I/var/target/usr/include/libfoo with
              a PKG_CONFIG_SYSROOT_DIR equal to /var/target (same rule
apply to -L)
"""

Cross Compiling Gnokii for ARM

Forconi — 2012-05-08T12:14:22

Hi,
I'm trying to cross compile Gnokii for an ARM platform to use smsd deamon with a serial GSM/GPRS modem.

The configure command that I've used is the following:
GLIB_CFLAGS="-I$ROOTFS_PATH/usr/include/glib-2.0 -I$ROOTFS_PATH/usr/lib/glib-2.0/include" \
GLIB_LIBS="-L$ROOTFS_PATH/usr/lib" \
./configure \
--host=arm-none-linux-gnueabi \
--prefix=/home/samuele/geam/gnokii/build \
--disable-postgres \
--disable-mysql \
--disable-sqlite \
--disable-libusb \
--disable-irda \
--disable-bluetooth \
--disable-libical \
--disable-phonet \
--disable-libpcsclite \
--disable-rlpdebug \
--disable-xdebug \
--disable-fulldebug \
--disable-nls \
--without-x

Before running this I exported the ROOTFS_PATH variable: 
export ROOTFS_PATH=/media/vol/geam/rootfs/armle-fs

This is the output of the configure script:
    Host system:        linux-gnueabi
    Gnokii version:     0.6.31
    Xgnokii version:    1.0
    SMSD version:       1.5.2
    X (GTK+) support:   no
    smsd support:       yes
    Postgres support:   no
    SQLi

Re: Forwarding SMS & VoIP gateway

Axel Gruber — 2012-05-02T15:19:33

Hi

i´m using some perl-Scripts to do this things:

- Forward incomming SMS with EMAIL
- Sort incomming SMS into different MYSQL Tables

If you want do do this you have to use the MYSQL Feature - all other is a
little bit of Perl Scripting - i think also forwad to a other SMS number is
no Problem at all because you ouly have to write it to the outbox table...



2012/5/2 Jago Pearce <jago25_98< at >hotmail.com>

_______________________________________________
gnokii-users mailing list
gnokii-users< at >nongnu.org
https://lists.nongnu.org/mailman/listinfo/gnokii-users

Forwarding SMS & VoIP gateway

Jago Pearce — 2012-05-02T14:38:18

 I just wondered if it might be possible to recieve a SMS on one phone and resend it on another?

Also, what about receiving a VoIP call via a linux client and then routing that and dialling out with the cell phone?

       _______________________________________________
gnokii-users mailing list
gnokii-users< at >nongnu.org
https://lists.nongnu.org/mailman/listinfo/gnokii-users

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Daniele Forsi — 2012-04-26T19:15:34

2012/4/26 Charles:


great
BTW, I was using a broken udev rule, the suggested one works for me


you have done your part


with the next release of libgnokii there will be no need to blacklist
cdc_phonet and phonet, you just need the write permissions set with
that udev rule

distributions can choose between two groups: the one used for
removable devices (eg. plugdev) or the one used for modems (eg.
dialout)

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Charles — 2012-04-26T14:51:28

The new version does as you intended:
c< at >CW8:~$ gnokii --identify
GNOKII Version 0.6.32git
LOG: debug mask is 0x1
Config read from file /home/c/.config/gnokii/config.
phone instance config:
model = series40
port = 1
connection = dku2libusb
initlength = default
serial_baudrate = 19200
serial_write_usleep = -1
handshake = software
require_dcd = 0
smsc_timeout = 10
rfcomm_channel = 0
sm_retry = 0
Connecting
overrun: 1010 0
device: opening device USB
Ignoring extra header, type 171, length 5
Found FBUS interface
Couldn't open PHONET device: Permission denied
Error in link initialisation: 1
Telephone interface init failed: Command failed.
Quitting.
Command failed.

FYI the previous Slackware installation did not need the gnokii udev
rule because VirtualBox was installed including a udev rule to set USB
devices writeable for group vboxusers.  The version of VirtualBox on the
new installation does not install a udev rule.

After creating /etc/udev/rules.d/45-nokiadku2.rules with the following
simplified content (th

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Daniele Forsi — 2012-04-26T12:32:51

2012/4/26 Charles:


doh, in retrospect that is the reason why it worked as root
can you try to update gnokii from git?
I just added a missing check for usb_open() failure, it should print
some error messages starting with:
Couldn't open PHONET device: Operation not permitted
then quit without a segfault


yes, look at
http://wiki.gnokii.org/index.php/DKU2#Setting_permissions_for_libusb

however that udev rule doesn't seem to work with my new debian install
(kernel3.2.0-2-686-pae , udev 175-3.1)
Did they change (again) the format?

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Charles — 2012-04-26T09:38:36

c< at >CW8:~$ ldd /usr/bin/gnokii
linux-vdso.so.1 =>  (0x00007fff990c6000)
libgnokii.so.7 => /usr/lib64/libgnokii.so.7 (0x00007f02a4274000)
libXpm.so.4 => /usr/lib64/libXpm.so.4 (0x00007f02a4063000)
libX11.so.6 => /usr/lib64/libX11.so.6 (0x00007f02a3d22000)
libxcb.so.1 => /usr/lib64/libxcb.so.1 (0x00007f02a3b06000)
libXau.so.6 => /usr/lib64/libXau.so.6 (0x00007f02a3903000)
libXdmcp.so.6 => /usr/lib64/libXdmcp.so.6 (0x00007f02a36fd000)
libdl.so.2 => /lib64/libdl.so.2 (0x00007f02a34de000)
libbluetooth.so.3 => /usr/lib64/libbluetooth.so.3 (0x00007f02a32c3000)
libusb-0.1.so.4 => /usr/lib64/libusb-0.1.so.4 (0x00007f02a30bd000)
libusb-1.0.so.0 => /usr/lib64/libusb-1.0.so.0 (0x00007f02a2eb0000)
libical.so.0 => /usr/lib64/libical.so.0 (0x00007f02a2c53000)
libpthread.so.0 => /lib64/libpthread.so.0 (0x00007f02a2a34000)
libglib-2.0.so.0 => /usr/lib64/libglib-2.0.so.0 (0x00007f02a271d000)
librt.so.1 => /lib64/librt.so.1 (0x00007f02a2514000)
libreadline.so.5 => /usr/lib64/libreadline.so.5 (0x00007f02a22d5000)

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Charles — 2012-04-25T13:55:49

I don't have a Slackware 13.37 (32-bit) OS so, yes, I could do it but
would have to start by downloading the installation media which would
take a couple of days (will start that now).  Can we continue
investigating on the existing OS first?

The same phone worked on Slackware64 13.1 with gnokii built from the
gnokii-master source downloaded on 22jun11.  Slackware64 13.1 had
libusb-1.0.6 and libusb-compat-0.1.3 -- whereas Slackware64 13.37 has
libusb-1.0.8 and the same libusb-compat-0.1.3.

Sorry for not realising stdio.h was required -- that's basic C programming!

I am not confident about getting the as-installed gnokii to use libusb
with the added debugging.  Was this the correct technique?

root< at >CW8:~# sed -i '22i\#include <stdio.h>'
/tmp/libusb-1.0.8/libusb/descriptor.c
root< at >CW8:~# sed -i '708i\        printf("%d %d %d\\n", di, si,
tbuf[0]);' /tmp/libusb-1.0.8/libusb/descriptor.c
[inspected the modified source; it looked as intended]
root< at >CW8:~# cd /tmp/libusb-1.0.8 && make
root< at >CW8:~# ls -l /tmp/packa

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Daniele Forsi — 2012-04-25T14:40:13

2012/4/25 Charles:


sure


trying to compile libusb-1.0.6 should be faster than downloading the 32bit OS


change the library search path at runtime to look first in the
directory containing the new libusb-0.1.so.4 or what gnokii searches
(use ldd gnokii to see needed libraries)

looking at your output it could be something like:
LD_LIBRARY_PATH=/tmp/package-libusb/usr/lib64/ gnokii --identify

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Daniele Forsi — 2012-04-25T09:43:49

2012/4/25 Charles:


ah, so the differences are that you're using a 64bit build and libusb-compat

Any chance you can try the same phone and gnokii version(s) on a
computer with 32bit and/or libusb-0?


add this:
#include <stdio.h>

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Charles — 2012-04-25T04:45:51

Tried that but ther eis nu usb.c in the source.

The sources distributed with Slackware64 13.37 are:
325K May  5  2010 libusb-1.0.8.tar.bz2
246K Dec  1  2009 libusb-compat-0.1.3.tar.bz2
 322 Mar 25  2011 libusb.fix.debug.log.level.diff.gz

After running the standard build, I could not find usb.c:

root< at >CW8:~# find /tmp -name '*.c'
/tmp/libusb-1.0.8/examples/dpfp.c
/tmp/libusb-1.0.8/examples/dpfp_threaded.c
/tmp/libusb-1.0.8/examples/lsusb.c
/tmp/libusb-1.0.8/libusb/core.c
/tmp/libusb-1.0.8/libusb/descriptor.c
/tmp/libusb-1.0.8/libusb/io.c
/tmp/libusb-1.0.8/libusb/os/darwin_usb.c
/tmp/libusb-1.0.8/libusb/os/linux_usbfs.c
/tmp/libusb-1.0.8/libusb/sync.c
/tmp/libusb-compat-0.1.3/examples/lsusb.c
/tmp/libusb-compat-0.1.3/examples/testlibusb.c
/tmp/libusb-compat-0.1.3/libusb/core.c

Function usb_get_string_simple is defined in
libusb-compat-0.1.3/libusb/core.c:

root< at >CW8:~# for f in $( find /tmp -name '*.c' )
  do buf=$( grep usb_get_string_simple $f)
  [[ $buf ]] && echo $f:$'\n'"$buf"
done
/tmp/libusb-compat-0

Re: SMSD register RING (AT+CLIP)

Daniele Forsi — 2012-04-24T09:14:02

2012/4/24  <luca< at >...>:


you didn't say that but I guess you also want to receive SMS


I'm not sure if it fits in the scope of smsd
there are other events that could be notified to smsd (eg. loss of GSM
coverage, so that you know in advance that sending will fail) but this
would make smsd more complicated for an uncertain gain
some operators will send an alert SMS if you refuse the call pressing
the red button on the phone at the first ring, just like when your
phone is unreachable, this might be a flaw in their systems but it can
be useful for you as long as it lasts (I think I've seen this with
Tre)
if you don't need to react immediately you can read the MC memory
every couple of seconds to get the list of calls missed when the phone
was reachable


ok, mi trovi anche su irc in #gnokii su freenode

Re: SMSD register RING (AT+CLIP)

2012-04-24T08:03:55

eheh ok sorry...
I'm need to record a missing call on MySQL table inbox (with a specific
flag).
This allows me to manage start / stop availability of firefighters from a
php portal at no cost for a single firefighter.


ops...i'm using serial modem (Wavecom) =)
I'm working to integrate waitcall.c in smsd...

Ciao!
l.

ps: quando funziona invio il diff ;-)

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Daniele Forsi — 2012-04-23T22:18:08

2012/4/23 Charles:


the other place to check is in libusb
if you can compile it from sources with debug symbols, then gdb will
tell the exact line else you will need to add printf's in file usb.c
in function usb_get_string_simple()

eg. printf("%d %d %d\n", di, si, tbuf[0]); just inside the for loop
(line 261 in my copy):
  for (di = 0, si = 2; si < tbuf[0]; si += 2) {

Re: SMSD register RING (AT+CLIP)

Daniele Forsi — 2012-04-23T17:28:36

2012/4/20  <luca< at >...>:


no, unless you want to parse stderr, but you are telling a *solution*
without the *problem*

hint: some USB devices create two ports so for example you could use
/dev/ttyACM0 for smsd and /dev/ttyACM1 for an application such as
waitcall from gnokii-extras (however not all functionality is always
present on both devices)

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Charles — 2012-04-23T17:27:37

GNOKII Version 0.6.32git
LOG: debug mask is 0x1
Config read from file /home/c/.config/gnokii/config.
phone instance config:
model = series40
port = 1
connection = dku2libusb
initlength = default
serial_baudrate = 19200
serial_write_usleep = -1
handshake = software
require_dcd = 0
smsc_timeout = 10
rfcomm_channel = 0
sm_retry = 0
Connecting
overrun: 1010 0
device: opening device USB
Ignoring extra header, type 171, length 5
Found FBUS interface
current->device: 0x65e0a0
Segmentation fault

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Pawel Kot — 2012-04-22T17:06:36

Hi,

On Sun, Apr 22, 2012 at 18:58, Charles <gnokii< at >catcons.co.uk> wrote:

So that looks good. Maybe:
dprintf("current->device: %p\n", current->device);

take care,

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Charles — 2012-04-22T16:58:51

Found FBUS interface
devinstance: 0x7fded6c43518
current: 0x65e5e0
Segmentation fault

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Pawel Kot — 2012-04-21T17:05:31

Hi,

On Sat, Apr 21, 2012 at 18:55, Charles <gnokii< at >catcons.co.uk> wrote:

Try adding the following before the first get_iface_string():
dprintf("devinstance: %p\n", &DEVINSTANCE(state));
dprintf("current: %p\n", current);
and show the outcome.

thanks,

Re: gnokii 0.6.29 and 0.6.32 master segfaults

Charles — 2012-04-21T16:55:31

dprintf() looked complex (!) so I used printf and found the segfault is
happening on the first call to get_iface_string (the printf before it
did, the one after did not).


0.1.12