Re: Driver for 536ep-Modem does not compile

Wolfgang Junker <junker@xxxxxxxxxxxxxxxxxxx> · Sun, 19 Feb 2012 19:07:59 +0100 (CET)

Dear Philippe,

here is the output of your test-script:

linux:~/Installationen/Modem-536ep/b # sh test.sh
#!/bin/sh > /etc/init.d//536EP
### BEGIN INIT INFO >> /etc/init.d//536EP
insserv: warning: script 'S01Intel536_boot' missing LSB tags and 
overrides
insserv: warning: script 'K01vpnclient_init' missing LSB tags and 
overrides
insserv: warning: script 'Intel536_boot' missing LSB tags and 
overrides
insserv: warning: script 'Intel536_boot' missing LSB tags and 
overrides
insserv: warning: script 'vpnclient_init' missing LSB tags and 
overrides
test.sh: Zeile 19: EOF: Kommando nicht gefunden.
linux:~/Installationen/Modem-536ep/b # echo $?
127

...................................................................

When I run wvdial I get the following messages:

--> WvDial: Internet dialer version 1.60
--> Cannot get information for serial port.
--> Initializing modem.
--> Sending: ATZ
ATZ
OK
--> Sending: ATQ0 V1 E1 S0=0 &C1 &D2
ATQ0 V1 E1 S0=0 &C1 &D2
OK
--> Modem initialized.
--> Idle Seconds = 300, disabling automatic reconnect.
--> Sending: ATDT01920787
--> Waiting for carrier.
ATDT01920787
CONNECT 50666
--> Carrier detected.  Starting PPP immediately.
--> Starting pppd at Sun Feb 19 16:24:44 2012
--> Pid of pppd: 14966
--> Using interface ppp0
--> pppd: [18][06][08]
--> pppd: [18][06][08]
--> pppd: [18][06][08]
--> pppd: [18][06][08]
--> local  IP address 145.254.249.173
--> pppd: [18][06][08]
--> remote IP address 145.253.1.232
--> pppd: [18][06][08]
--> Script /etc/ppp/ip-up run successful
--> Default route Ok.
--> warning, can't find address for `www.suse.de`
--> warning, address lookup does not work
--> Nameserver (DNS) failure, the connection may not work.
--> Connected... Press Ctrl-C to disconnect
--> pppd: [18][06][08]
.................................................................

and $ tail -f /var/log/messages gives:

Feb 19 16:24:44 linux pppd[14966]: Plugin passwordfd.so loaded.
Feb 19 16:24:44 linux pppd[14966]: pppd 2.4.5 started by root, 
uid 0
Feb 19 16:24:44 linux pppd[14966]: Using interface ppp0
Feb 19 16:24:44 linux pppd[14966]: Connect: ppp0 <--> /dev/536ep
Feb 19 16:24:44 linux modem-manager: (net/ppp0): could not get 
port's parent device
Feb 19 16:24:45 linux pppd[14966]: CHAP authentication succeeded
Feb 19 16:24:45 linux pppd[14966]: CHAP authentication succeeded
Feb 19 16:24:45 linux pppd[14966]: local  IP address 
145.254.249.173
Feb 19 16:24:45 linux pppd[14966]: remote IP address 
145.253.1.232
Feb 19 16:24:46 linux pppd[14966]: Script /etc/ppp/ip-up finished 
(pid 14968), status = 0x0
..................................................................

The Config-file /etc/wvdial.conf looks like this:

[Dialer Defaults]
Modem = /dev/modem
Baud = 57600
Init1 = ATZ
Init2 = ATQ0 V1 E1 S0=0 &C1 &D2
Init3 =
Area Code =
Phone = 01920787
Username = arcor
Password = internet
Ask Password = 0
Dial Command = ATDT
Stupid Mode = 1
Compuserve = 0
Force Address =
Idle Seconds = 300
DialMessage1 =
DialMessage2 =
ISDN = 0
Auto DNS = 0
....................................................................

It seems that SUSE has adapted wvdial, because I didn't look for 
'www.suse.de', this seems to be a built-in DNS-check of SUSE. The 
option "Auto DNS = " in the wvdial.conf is also due to SUSE and 
asks whether to get the DNS from the provider or not. I tried "0" 
and "1" but experienced the same behaviour with both choices.

Cordially
Wolfgang

On Sun, 19 Feb 2012, Philippe Vouters wrote:

Dear Wolfgang,

As a beginning to this reply and basing the /etc/init.d/Intel536_inst code 
from your stoppreload boot script, can try the following from a root account:

First commands:
# mkdir b
# cd b
# cp /path/to/intel-536-537/Intel536_boot ./
# cat > test.sh <<- EOF
#!bin/sh
INITDIR=/etc/init.d/
echo "#!/bin/sh >" ${INITDIR}/536EP
echo "### BEGIN INIT INFO >>" ${INITDIR}/536EP
echo "# Provides:          Intel536_boot" >> ${INITDIR}/536EP
echo "# Required-Start:    \$syslog" >> ${INITDIR}/536EP
echo "# Required-Stop:     \$syslog" >> ${INITDIR}/536EP
echo "# Default-Start:     3 5" >> ${INITDIR}/536EP
echo "# Default-Stop:      1 2" >> ${INITDIR}/536EP
echo "# Description:       Start/Stop the Intel 536EP driver" >> 
${INITDIR}/536EP
echo "### END INIT INFO" >> ${INITDIR}/536EP
sed '1,6d' Intel536_boot >  /tmp/Intel536_boot
cat ${INITDIR}/536EP /tmp/Intel536_boot > ${INITDIR}/Intel536_boot
chown root:root ${INITDIR}/Intel536_boot
chmod 755 ${INITDIR}/Intel536_boot
insserv ${INITDIR}/Intel536_boot
rm ${INITDIR}/536EP
rm /tmp/Intel536_boot
EOF

Finally:

sh test.sh
echo $?

efax sounds quite excellent with absolutely nothing abnormal in 
/var/log/messages. wvdial sound much better. For wvdial, I think you should 
have a closer look at your /etc/wvdial.conf. I would need your wvdial run 
output to better understand the problem you are facing. My guess is that pppd 
does not correctly setup your Internet connection.

Philippe

Le 18/02/2012 20:02, Wolfgang Junker a écrit :
Dear Philippe,

sorry for my misunderstanding, here is the file /etc/init.d/stoppreload:

#! /bin/sh
#
# Copyright (c) 2007 SuSE Linux Products GmbH Nuernberg, Germany. All
rights reserved.
#
### BEGIN INIT INFO
# Provides: stoppreload
# Required-Start: $ALL boot.startpreload $remote_fs
# Required-Stop: $null
# Default-Start: 1 2 3 5
# Default-Stop:
# Description: Stopping and parsing the preloadtrace output
# Short-Description: Stop preload (and blktrace)
### END INIT INFO

. /etc/rc.status
rc_reset

case "$1" in
start)
/usr/sbin/stop_preload &
rc_status -v -r
;;
stop)
rc_status -v
;;
reload)
rc_failed 3
rc_status -v
;;
status)
rc_failed 4
rc_status -v
;;
*)
echo "Usage: $0 {start|stop|status}"
exit 1
;;
esac

rc_exit
.....................................................................

I compiled the efax and did $./efax -vewinchmart, the output is as follows:

./efax: Sat Feb 18 17:18:16 2012 efax v 0.9a-001114 Copyright 1999 Ed Casas
./efax: Sat Feb 18 17:18:16 2012 efax v 0.9a-001114 Copyright 1999 Ed Casas
efax: 18:16 compiled Feb 16 2012 22:50:31
efax: 18:16 argv[0]=./efax
efax: 18:16 argv[1]=-vewinchmart
efax: 18:16 using built-in 128x0 font
efax: 18:18 opened /dev/modem
efax: 18:18 command "Q0V1"
efax: 18:18 waiting 2.0 s
efax: 18:28 command "Q0V1"
efax: 18:28 waiting 2.0 s
efax: 18:28 .765 [ATQ0V1<CR><CR><LF>]
efax: 18:28 .829 [OK<CR><LF>]
efax: 18:28 response "OK"
efax: 18:28 command "E0"
efax: 18:28 waiting 5.0 s
efax: 18:28 .849 [ATE0<CR><CR><LF>]
efax: 18:28 .896 [OK<CR><LF>]
efax: 18:28 response "OK"
efax: 18:28 command "I3"
efax: 18:28 waiting 5.0 s
efax: 18:28 .948 [<CR><LF>536EP<CR><LF>]
efax: 18:28 .969 [<CR><LF>OK<CR><LF>]
efax: 18:28 response "OK"
efax: 18:28 command "+FCLASS=?"
efax: 18:28 waiting 5.0 s
efax: 18:29 .016 [<CR><LF>0,1,8<CR><LF>]
efax: 18:29 .037 [<CR><LF>OK<CR><LF>]
efax: 18:29 response "OK"
efax: 18:29 command "+FCLASS=1"
efax: 18:29 waiting 5.0 s
efax: 18:29 .108 [<CR><LF>OK<CR><LF>]
efax: 18:29 response "OK"
efax: 18:29 using 536EP in class 1
efax: 18:29 command "+FRM=?"
efax: 18:29 waiting 5.0 s
efax: 18:29 .157 [<CR><LF>24,48,72,73,74,96,97,98,121,122,145,146<CR><LF>]
efax: 18:29 .177 [<CR><LF>OK<CR><LF>]
efax: 18:29 response "OK"
efax: 18:29 command "A"
efax: 18:29 waiting 120.0 s
efax: 18:34 .068 [<CR><LF>CONNECT<CR><LF>]
efax: 18:34 response "CONNECT"
efax: 18:34 fax call answered
efax: 18:34 sent 23 bytes:
efax: 18:34 ff c0 02 04 04 04 04 04 04 04 04 04 04 04 04 04
efax: 18:34 04 04 04 04 04 04 04
efax: 18:34 sent CSI - answering ID
efax: 18:34 waiting 13.6 s
efax: 18:35 .796 [<CR><LF>CONNECT<CR><LF>]
efax: 18:35 response "CONNECT"
efax: 18:35 sent 6 bytes: ff c8 01 00 76 1e
efax: 18:35 sent DIS - answering capabilities
efax: 18:35 waiting 13.6 s
efax: 18:36 .156 [<CR><LF>OK<CR><LF>]
efax: 18:36 response "OK"
efax: 18:36 command "+FRH=3"
efax: 18:36 waiting 3.0 s
efax: 18:39 waiting 2.0 s
efax: 18:39 .189 [<CR><LF>OK<CR><LF>]
efax: 18:39 response "OK"
efax: 18:39 received 0 bytes:
efax: 18:39 received DIS - answering capabilities
efax: 18:39 command "+FTH=3"
efax: 18:39 waiting 3.1 s
efax: 18:39 .257 [<CR><LF>CONNECT<CR><LF>]
efax: 18:39 response "CONNECT"
efax: 18:39 sent 23 bytes:
efax: 18:39 ff c0 02 04 04 04 04 04 04 04 04 04 04 04 04 04
efax: 18:39 04 04 04 04 04 04 04
efax: 18:39 sent CSI - answering ID
efax: 18:39 waiting 13.6 s
efax: 18:40 .988 [<CR><LF>CONNECT<CR><LF>]
efax: 18:40 response "CONNECT"
efax: 18:40 sent 6 bytes: ff c8 01 00 76 1e
efax: 18:40 sent DIS - answering capabilities
efax: 18:40 waiting 13.6 s
efax: 18:41 .348 [<CR><LF>OK<CR><LF>]
efax: 18:41 response "OK"
efax: 18:41 command "+FRH=3"
efax: 18:41 waiting 3.0 s
efax: 18:44 waiting 2.0 s
efax: 18:44 .377 [<CR><LF>OK<CR><LF>]
efax: 18:44 response "OK"
efax: 18:44 received 0 bytes:
efax: 18:44 received DIS - answering capabilities
efax: 18:44 command "+FTH=3"
efax: 18:44 waiting 3.1 s
efax: 18:44 .448 [<CR><LF>CONNECT<CR><LF>]
efax: 18:44 response "CONNECT"
efax: 18:44 sent 23 bytes:
efax: 18:44 ff c0 02 04 04 04 04 04 04 04 04 04 04 04 04 04
efax: 18:44 04 04 04 04 04 04 04
efax: 18:44 sent CSI - answering ID
efax: 18:44 waiting 13.6 s
efax: 18:46 .177 [<CR><LF>CONNECT<CR><LF>]
efax: 18:46 response "CONNECT"
efax: 18:46 sent 6 bytes: ff c8 01 00 76 1e
efax: 18:46 sent DIS - answering capabilities
efax: 18:46 waiting 13.6 s
efax: 18:46 .536 [<CR><LF>OK<CR><LF>]
efax: 18:46 response "OK"
efax: 18:46 command "+FRH=3"
efax: 18:46 waiting 3.0 s
efax: 18:49 waiting 2.0 s
efax: 18:49 .569 [<CR><LF>OK<CR><LF>]
efax: 18:49 response "OK"
efax: 18:49 received 0 bytes:
efax: 18:49 received DIS - answering capabilities
efax: 18:49 command "+FTH=3"
efax: 18:49 waiting 3.1 s
efax: 18:49 .636 [<CR><LF>CONNECT<CR><LF>]
efax: 18:49 response "CONNECT"
efax: 18:49 sent 23 bytes:
efax: 18:49 ff c0 02 04 04 04 04 04 04 04 04 04 04 04 04 04
efax: 18:49 04 04 04 04 04 04 04
efax: 18:49 sent CSI - answering ID
efax: 18:49 waiting 13.6 s
efax: 18:51 .368 [<CR><LF>CONNECT<CR><LF>]
efax: 18:51 response "CONNECT"
efax: 18:51 sent 6 bytes: ff c8 01 00 76 1e
efax: 18:51 sent DIS - answering capabilities
efax: 18:51 waiting 13.6 s
efax: 18:51 .728 [<CR><LF>OK<CR><LF>]
efax: 18:51 response "OK"
efax: 18:51 command "+FRH=3"
efax: 18:51 waiting 3.0 s
efax: 18:54 waiting 2.0 s
efax: 18:54 .758 [<CR><LF>OK<CR><LF>]
efax: 18:54 response "OK"
efax: 18:54 received 0 bytes:
efax: 18:54 received DIS - answering capabilities
efax: 18:54 command "+FTH=3"
efax: 18:54 waiting 3.1 s
efax: 18:54 .828 [<CR><LF>CONNECT<CR><LF>]
efax: 18:54 response "CONNECT"
efax: 18:54 sent 23 bytes:
efax: 18:54 ff c0 02 04 04 04 04 04 04 04 04 04 04 04 04 04
efax: 18:54 04 04 04 04 04 04 04
efax: 18:54 sent CSI - answering ID
efax: 18:54 waiting 13.6 s
efax: 18:56 .557 [<CR><LF>CONNECT<CR><LF>]
efax: 18:56 response "CONNECT"
efax: 18:56 sent 6 bytes: ff c8 01 00 76 1e
efax: 18:56 sent DIS - answering capabilities
efax: 18:56 waiting 13.6 s
efax: 18:56 .916 [<CR><LF>OK<CR><LF>]
efax: 18:56 response "OK"
efax: 18:56 command "+FRH=3"
efax: 18:56 waiting 3.0 s
efax: 18:59 waiting 2.0 s
efax: 18:59 .948 [<CR><LF>OK<CR><LF>]
efax: 18:59 response "OK"
efax: 18:59 received 0 bytes:
efax: 18:59 received DIS - answering capabilities
efax: 18:59 Error: no command/response from remote
efax: 18:59 command "+FTH=3"
efax: 18:59 waiting 3.1 s
efax: 19:00 .016 [<CR><LF>CONNECT<CR><LF>]
efax: 19:00 response "CONNECT"
efax: 19:00 sent 3 bytes: ff c8 5f
efax: 19:00 sent DCN - disconnect
efax: 19:00 waiting 13.6 s
efax: 19:01 .256 [<CR><LF>OK<CR><LF>]
efax: 19:01 response "OK"
efax: 19:01 command "H"
efax: 19:01 waiting 5.0 s
efax: 19:01 .708 [<CR><LF>OK<CR><LF>]
efax: 19:01 response "OK"
efax: 19:01 command "Q0V1"
efax: 19:01 waiting 2.0 s
efax: 19:01 .788 [<CR><LF>OK<CR><LF>]
efax: 19:01 response "OK"
efax: 19:01 command "H"
efax: 19:01 waiting 5.0 s
efax: 19:02 .328 [<CR><LF>OK<CR><LF>]
efax: 19:02 response "OK"
efax: 19:02 done, returning 3 (invalid modem response)
....................................................................

tail -f /var/log/messages gives:

Feb 18 17:16:16 linux su: (to root) junker on /dev/pts/4
Feb 18 17:18:16 linux kernel: [ 3785.035276] pci 0000:00:06.0: enabling
device (0010 -> 0012)
Feb 18 17:18:16 linux kernel: [ 3785.035294] pci 0000:00:06.0: PCI INT A
-> Link[LNKA] -> GSI 10 (level, low) -> IRQ 10

....................................................................

This looked quite pretty to my, so I tried wvdial (not wvdialconf, which
looks for the wrong serial ports), and in fact I get a stable connection
to an Internet Provider of my choice!!
The bad news is, there is no throughput, wvdial says the following:

--> warning, address lookup does not work
--> Nameserver (DNS) failure, the connection may not work.

This seems to be a new issue or is it connected to anything before?
Anyway, if you think at some point it's useful to ssh to my computer, I
wouldn't mind.

Sincerely
Wolfgang

On Fri, 17 Feb 2012, Philippe Vouters wrote:

Dear Wolgang,

Yes indeed I was a bit surprised that in a so population dense region
such as Aachen and in a so developed country such as Germany you did
not have a broadband facility. But I played your game.

Perhaps, I poorly expressed myself. The build
/etc/init.d/Intel536_boot and /etc/init.d/vpnclient_init you just sent
me are NOT fully suited to a modern SUSE distribution such as yours as
it showed up with $ sudo make install with the "set -x" you added on
my demand. These two boot scripts show errors when installing them.
Can you send me any other one but these two boot scripts you will find
in /etc/init.d ?

Why do I insist on running efax first ? This is because it is a much
simpler software using the modem only involving itself. It is much
more easy to understand what can be going wrong with the driver's
installation and anyhow and above all if efax does not run correctly,
there is strictly no hope that wvdial will correctly work. In
conclusion, efax with its simplicity is an ideal candidate to
troubleshoot any modem driver.

Note that for the runs below I run a special driver that I modified
for one user also running a Fedora 16 distribution. The previous
version showed up one kernel oops under kernel 3.1.x that I never
noticed under 2.6.x kernels. So I keep this in my mind, just in case
my changes need more work. I just commented out parts of the code in a
problem reproduced 536EP run using efax. I am not yet fully sure about
the code parts I can safely recomment in. As you have a DSL
subscription, this oops and its temporary solution is described at
http://vouters.dyndns.org/tima/Linux-drivers-Troubleshooting_a_oops.html
for your curiosity.

Read hereafter my today's run under Linux Fedora 16 (kernel 3.2.6). If
you do not read such, we shall adjust for our next mail exchanges.

Also as I feel you are confident in me and provided we are really
short of ideas, would you authorize me with much care against possible
unwanted intruders to SSH connect to your SUSE system ? I already did
it for a modem issue connecting to a computer somewhere in Latin
America. I have been remotely working on the system while its owner
who was so tired was quietly sleeping. When he woke up, he noticed a
mail from me telling his modem was perfectly working. I got such a
warm thank that I'll never forget it.

With my warmest regards to you.
Philippe

[philippe@victor ~]$ uname -r
3.2.6-3.fc16.i686
[philippe@victor intel-536-537]$ sudo make install
make[1]: Entering directory `/home/philippe/intel-536-537'
rm -f /etc/hamregistry.bin
bash Intel536_inst
running kernel 3.2.6-3.fc16.i686
installing hamregistry, used for persistant storage
installing Intel536 driver
install REDHAT Intel536 boot script and links
starting module and utilities
hamregistry: no process found

done
make[1]: Leaving directory `/home/philippe/intel-536-537'

[philippe@victor intel-536-537]$ dmesg | tail
...
[ 5134.776989] Intel 536EP card found
...

Now running efax:

[philippe@victor ~]$ efax -vewinchmart
efax: Fri Feb 17 22:15:24 2012 efax v 0.9a-001114 Copyright 1999 Ed Casas
efax: Fri Feb 17 22:15:24 2012 efax v 0.9a-001114 Copyright 1999 Ed Casas
efax: 15:24 compiled Feb 8 2011 17:28:56
efax: 15:24 argv[0]=efax
efax: 15:24 argv[1]=-vewinchmart
efax: 15:25 opened /dev/modem
efax: 15:25 command "Q0V1"
efax: 15:25 waiting 2.0 s
efax: 15:28 waiting 2.0 s
efax: 15:28 .494 [ATQ0V1<CR><CR><LF>]
efax: 15:28 .554 [OK<CR><LF>]
efax: 15:28 response "OK"
efax: 15:28 waiting 2.0 s
efax: 15:30 command "Q0V1"
efax: 15:30 waiting 2.0 s
efax: 15:30 .674 [ATQ0V1<CR><CR><LF>]
efax: 15:30 .734 [OK<CR><LF>]
efax: 15:30 response "OK"
efax: 15:30 command "E0"
efax: 15:30 waiting 5.0 s
efax: 15:30 .854 [ATE0<CR><CR><LF>]
efax: 15:30 .904 [OK<CR><LF>]
efax: 15:30 response "OK"
efax: 15:31 command "I3"
efax: 15:31 waiting 5.0 s
efax: 15:31 .054 [<CR><LF>536EP<CR><LF>]
efax: 15:31 .074 [<CR><LF>OK<CR><LF>]
efax: 15:31 response "OK"
efax: 15:31 command "+FCLASS=?"
efax: 15:31 waiting 5.0 s
efax: 15:31 .224 [<CR><LF>0,1,8<CR><LF>]
efax: 15:31 .244 [<CR><LF>OK<CR><LF>]
efax: 15:31 response "OK"
efax: 15:31 command "+FCLASS=1"
efax: 15:31 waiting 5.0 s
efax: 15:31 .414 [<CR><LF>OK<CR><LF>]
efax: 15:31 response "OK"
efax: 15:31 using 536EP in class 1
efax: 15:31 command "+FRM=?"
efax: 15:31 waiting 5.0 s
efax: 15:31 .564
[<CR><LF>24,48,72,73,74,96,97,98,121,122,145,146<CR><LF>]
efax: 15:31 .584 [<CR><LF>OK<CR><LF>]
efax: 15:31 response "OK"
efax: 15:31 command "A"
efax: 15:31 waiting 120.0 s
efax: 15:36 .504 [<CR><LF>CONNECT<CR><LF>]
efax: 15:36 response "CONNECT"
efax: 15:36 fax call answered
efax: 15:36 sent 23 bytes:
efax: 15:36 ff c0 02 04 04 04 04 04 04 04 04 04 04 04 04 04
efax: 15:36 04 04 04 04 04 04 04
efax: 15:36 sent CSI - answering ID
efax: 15:36 waiting 13.6 s
efax: 15:38 .234 [<CR><LF>CONNECT<CR><LF>]
efax: 15:38 response "CONNECT"
efax: 15:38 sent 6 bytes: ff c8 01 00 76 1e
efax: 15:38 sent DIS - answering capabilities
efax: 15:38 waiting 13.6 s
efax: 15:38 .594 [<CR><LF>OK<CR><LF>]
efax: 15:38 response "OK"
efax: 15:38 command "+FRH=3"
efax: 15:38 waiting 3.0 s
efax: 15:41 waiting 2.0 s
efax: 15:41 .624 [<CR><LF>OK<CR><LF>]
efax: 15:41 response "OK"
efax: 15:41 received 79 bytes:
efax: 15:41 ff c8 01 00 76 1e 04 04 04 04 04 04 04 04 04 04
efax: 15:41 04 04 04 04 04 04 04 00 00 00 dd 01 00 00 00 00
efax: 15:41 00 00 f4 cf 33 36 60 07 fb 43 44 6e ee bf 01 00
efax: 15:41 00 00 f4 cf fa 43 87 34 ef bf 87 34 ef bf b8 3a
efax: 15:41 ef bf 04 57 eb 43 35 5b 05 08 60 07 fb 43 3f
efax: 15:41 received DIS - answering capabilities
efax: 15:41 command "+FTH=3"
efax: 15:41 waiting 3.1 s
efax: 15:41 .694 [<CR><LF>CONNECT<CR><LF>]
efax: 15:41 response "CONNECT"
efax: 15:41 sent 23 bytes:
efax: 15:41 ff c0 02 04 04 04 04 04 04 04 04 04 04 04 04 04
efax: 15:41 04 04 04 04 04 04 04
efax: 15:41 sent CSI - answering ID
efax: 15:41 waiting 13.6 s
efax: 15:43 .424 [<CR><LF>CONNECT<CR><LF>]
efax: 15:43 response "CONNECT"
efax: 15:43 sent 6 bytes: ff c8 01 00 76 1e
efax: 15:43 sent DIS - answering capabilities
efax: 15:43 waiting 13.6 s
efax: 15:43 .784 [<CR><LF>OK<CR><LF>]
efax: 15:43 response "OK"
efax: 15:43 command "+FRH=3"
efax: 15:43 waiting 3.0 s
efax: 15:46 waiting 2.0 s
efax: 15:46 .814 [<CR><LF>OK<CR><LF>]
efax: 15:46 response "OK"
efax: 15:46 received 79 bytes:
efax: 15:46 ff c8 01 00 76 1e 04 04 04 04 04 04 04 04 04 04
efax: 15:46 04 04 04 04 04 04 04 00 00 00 dd 01 00 00 00 00
efax: 15:46 00 00 f4 cf 33 36 60 07 fb 43 44 6e ee bf 01 00
efax: 15:46 00 00 f4 cf fa 43 87 34 ef bf 87 34 ef bf b8 3a
efax: 15:46 ef bf 04 57 eb 43 35 5b 05 08 60 07 fb 43 3f
efax: 15:46 received DIS - answering capabilities
efax: 15:46 command "+FTH=3"
efax: 15:46 waiting 3.1 s
efax: 15:46 .884 [<CR><LF>CONNECT<CR><LF>]
efax: 15:46 response "CONNECT"
efax: 15:46 sent 23 bytes:
efax: 15:46 ff c0 02 04 04 04 04 04 04 04 04 04 04 04 04 04
efax: 15:46 04 04 04 04 04 04 04
efax: 15:46 sent CSI - answering ID
efax: 15:46 waiting 13.6 s
efax: 15:48 .614 [<CR><LF>CONNECT<CR><LF>]
efax: 15:48 response "CONNECT"
efax: 15:48 sent 6 bytes: ff c8 01 00 76 1e
efax: 15:48 sent DIS - answering capabilities
efax: 15:48 waiting 13.6 s
efax: 15:48 .974 [<CR><LF>OK<CR><LF>]
efax: 15:48 response "OK"
efax: 15:48 command "+FRH=3"
efax: 15:48 waiting 3.0 s
efax: 15:51 waiting 2.0 s
efax: 15:52 .004 [<CR><LF>OK<CR><LF>]
efax: 15:52 response "OK"
efax: 15:52 received 79 bytes:
efax: 15:52 ff c8 01 00 76 1e 04 04 04 04 04 04 04 04 04 04
efax: 15:52 04 04 04 04 04 04 04 00 00 00 dd 01 00 00 00 00
efax: 15:52 00 00 f4 cf 33 36 60 07 fb 43 44 6e ee bf 01 00
efax: 15:52 00 00 f4 cf fa 43 87 34 ef bf 87 34 ef bf b8 3a
efax: 15:52 ef bf 04 57 eb 43 35 5b 05 08 60 07 fb 43 3f
efax: 15:52 received DIS - answering capabilities
efax: 15:52 command "+FTH=3"
efax: 15:52 waiting 3.1 s
efax: 15:52 .074 [<CR><LF>CONNECT<CR><LF>]
efax: 15:52 response "CONNECT"
efax: 15:52 sent 23 bytes:
efax: 15:52 ff c0 02 04 04 04 04 04 04 04 04 04 04 04 04 04
efax: 15:52 04 04 04 04 04 04 04
efax: 15:52 sent CSI - answering ID
efax: 15:52 waiting 13.6 s
efax: 15:53 .804 [<CR><LF>CONNECT<CR><LF>]
efax: 15:53 response "CONNECT"
efax: 15:53 sent 6 bytes: ff c8 01 00 76 1e
efax: 15:53 sent DIS - answering capabilities
efax: 15:53 waiting 13.6 s
efax: 15:54 .164 [<CR><LF>OK<CR><LF>]
efax: 15:54 response "OK"
efax: 15:54 command "+FRH=3"
efax: 15:54 waiting 3.0 s
efax: 15:57 waiting 2.0 s
efax: 15:57 .194 [<CR><LF>OK<CR><LF>]
efax: 15:57 response "OK"
efax: 15:57 received 79 bytes:
efax: 15:57 ff c8 01 00 76 1e 04 04 04 04 04 04 04 04 04 04
efax: 15:57 04 04 04 04 04 04 04 00 00 00 dd 01 00 00 00 00
efax: 15:57 00 00 f4 cf 33 36 60 07 fb 43 44 6e ee bf 01 00
efax: 15:57 00 00 f4 cf fa 43 87 34 ef bf 87 34 ef bf b8 3a
efax: 15:57 ef bf 04 57 eb 43 35 5b 05 08 60 07 fb 43 3f
efax: 15:57 received DIS - answering capabilities
efax: 15:57 command "+FTH=3"
efax: 15:57 waiting 3.1 s
efax: 15:57 .264 [<CR><LF>CONNECT<CR><LF>]
efax: 15:57 response "CONNECT"
efax: 15:57 sent 23 bytes:
efax: 15:57 ff c0 02 04 04 04 04 04 04 04 04 04 04 04 04 04
efax: 15:57 04 04 04 04 04 04 04
efax: 15:57 sent CSI - answering ID
efax: 15:57 waiting 13.6 s
efax: 15:58 .994 [<CR><LF>CONNECT<CR><LF>]
efax: 15:58 response "CONNECT"
efax: 15:58 sent 6 bytes: ff c8 01 00 76 1e
efax: 15:58 sent DIS - answering capabilities
efax: 15:58 waiting 13.6 s
efax: 15:59 .354 [<CR><LF>OK<CR><LF>]
efax: 15:59 response "OK"
efax: 15:59 command "+FRH=3"
efax: 15:59 waiting 3.0 s
efax: 16:02 waiting 2.0 s
efax: 16:02 .384 [<CR><LF>OK<CR><LF>]
efax: 16:02 response "OK"
efax: 16:02 received 79 bytes:
efax: 16:02 ff c8 01 00 76 1e 04 04 04 04 04 04 04 04 04 04
efax: 16:02 04 04 04 04 04 04 04 00 00 00 dd 01 00 00 00 00
efax: 16:02 00 00 f4 cf 33 36 60 07 fb 43 44 6e ee bf 01 00
efax: 16:02 00 00 f4 cf fa 43 87 34 ef bf 87 34 ef bf b8 3a
efax: 16:02 ef bf 04 57 eb 43 35 5b 05 08 60 07 fb 43 3f
efax: 16:02 received DIS - answering capabilities
efax: 16:02 Error: no command/response from remote
efax: 16:02 command "+FTH=3"
efax: 16:02 waiting 3.1 s
efax: 16:02 .454 [<CR><LF>CONNECT<CR><LF>]
efax: 16:02 response "CONNECT"
efax: 16:02 sent 3 bytes: ff c8 5f
efax: 16:02 sent DCN - disconnect
efax: 16:02 waiting 13.6 s
efax: 16:03 .694 [<CR><LF>OK<CR><LF>]
efax: 16:03 response "OK"
efax: 16:03 command "H"
efax: 16:03 waiting 5.0 s
efax: 16:04 .144 [<CR><LF>OK<CR><LF>]
efax: 16:04 response "OK"
efax: 16:04 command "Q0V1"
efax: 16:04 waiting 2.0 s
efax: 16:04 .324 [<CR><LF>OK<CR><LF>]
efax: 16:04 response "OK"
efax: 16:04 command "H"
efax: 16:04 waiting 5.0 s
efax: 16:04 .864 [<CR><LF>OK<CR><LF>]
efax: 16:04 response "OK"
efax: 16:04 done, returning 3 (invalid modem response)
[philippe@victor ~]$

Checking with $ tail -f /var/log/messages

Feb 17 22:15:24 victor kernel: [ 5587.556264] type=1400
audit(1329513324.763:95): avc: denied { read write } for pid=9012
comm="efax" name="536ep" dev=devtmpfs ino=70965
scontext=user_u:user_r:user_t:s0 tcontext=user_u:object_r:device_t:s0
tclass=chr_file
Feb 17 22:15:24 victor kernel: [ 5587.556276] type=1400
audit(1329513324.763:96): avc: denied { open } for pid=9012
comm="efax" name="536ep" dev=devtmpfs ino=70965
scontext=user_u:user_r:user_t:s0 tcontext=user_u:object_r:device_t:s0
tclass=chr_file
Feb 17 22:15:24 victor kernel: [ 5587.556324] pci 0000:03:04.0: PCI
INT A -> GSI 17 (level, low) -> IRQ 17
Feb 17 22:15:24 victor kernel: [ 5587.556733] type=1400
audit(1329513324.763:97): avc: denied { write } for pid=8262
comm="hamregistry" name="ham" dev=proc ino=4026532245
scontext=user_u:user_r:user_t:s0 tcontext=system_u:object_r:proc_t:s0
tclass=file
Feb 17 22:15:25 victor kernel: [ 5588.057234] type=1400
audit(1329513325.264:98): avc: denied { ioctl } for pid=9012
comm="efax" path="/dev/536ep" dev=devtmpfs ino=70965
scontext=user_u:user_r:user_t:s0 tcontext=user_u:object_r:device_t:s0
tclass=chr_file

Le 17/02/2012 21:43, Wolfgang Junker a écrit :
Dear Philippe,

I have been using your modem drivers on this computer for almost ten
years, starting from my very first ancient Redhat-installation until my
previous SUSE 10.3-distribution. They have been invaluable to me and I
am much indebted to you for your
great job! Nowadays I am using DSL to connect to the internet, as
probably most of us do, and would like to use the modem for sending
faxes and as an alternative, if DSL is down. But not everybody in the
world is in such a comfortable position and I think it is really
worthwhile to keep the drivers up to date. After all, this is one of the
great things about Linux, that I can have the most uptodate and
excellent system on a 10 year old computer (but with good hardware...).
If I tried to put Windows7 on my computer it wouldn't make a peep...
Therefore I am happy that I can assist you in improving the code, and
after all, it's fun to solve a problem and learn a little bit more about
linux.

Having said this let us come back to our problem:

Here is the vpnclient_init script from /etc/init.d:

#!/bin/sh
##########################################################################

# Copyright (c) 2001, Cisco Systems, All Rights Reserved
###########################################################################

#
# File: vpnclient_init
# Date: 04/23/2001
#
###########################################################################

#
# chkconfig: 345 85 85
# description: Startup script for the vpn client. Version 4.8.02 (0030)
#
###########################################################################

# Source function library.
VPNCLIENT="/opt/cisco-vpnclient/bin/vpnclient"
VPNDEV="cipsec0"
VPNMOD=cisco_ipsec
case `uname -r` in
2.[56].*)
VPNMOD_FILE="${VPNMOD}.ko"
;;
*)
VPNMOD_FILE="$VPNMOD"
;;
esac

WHOAMI=`id | sed -e 's/(.*//'`

# See how we were called.
case "$1" in
start)
echo -n "Starting ${VPNCLIENT}: "
if [ "$WHOAMI" != "uid=0" ] ; then
echo "Failed (super user access required)"
exit 1
fi
/sbin/lsmod | grep -q "${VPNMOD}"
if [ "$?" = "0" ] ; then
echo "module ${VPNMOD} is already running. Use restart instead."
exit 1
fi

if [ -f /etc/resolv.conf.vpnbackup ]; then
echo "restoring /etc/resolv.conf"
mv /etc/resolv.conf.vpnbackup /etc/resolv.conf
fi
if [ -d /lib/modules/preferred ]; then
PC=/lib/modules/preferred/CiscoVPN
else
PC=/lib/modules/`uname -r`/CiscoVPN
fi
if [ -d $PC ] ; then
/sbin/insmod ${PC}/${VPNMOD_FILE}
if [ "$?" != "0" ] ; then
echo "Failed (insmod)"
exit 1
fi
else
echo "module directory $PC not found."
exit 1
fi
case "`uname -r`" in
2.6.*)
;;
2.5.*)
;;
2.4.*)
;;
2.2.*)
;;
2.0.*)
#
# This is only needed due to a bug in 2.0.x kernels that affects
# arp lookups.
#
ifconfig $VPNDEV 222.222.222.222 ;
if [ "$?" != "0" ] ; then
echo "Failed (ifconfig)"
/sbin/rmmod ${VPNMOD}
exit 1
fi
;;
*)
echo "Failed (unsupported Linux version)"
/sbin/rmmod ${VPNMOD}
exit 1
;;
esac

echo "Done"
;;
stop)
echo -n "Shutting down ${VPNCLIENT}: "
if [ "$WHOAMI" != "uid=0" ] ; then
echo "Failed (super user access required)"
exit 1
fi
killall cvpnd > /dev/null 2>&1

/sbin/lsmod | grep -q "${VPNMOD}"
if [ "$?" != "0" ] ; then
echo "module ${VPNMOD} is not running."
exit 1
fi
/sbin/ifconfig $VPNDEV down
if [ "$?" != "0" ] ; then
echo "Failed (ifconfig)"
exit 1
fi
/sbin/rmmod ${VPNMOD}
if [ "$?" != "0" ] ; then
echo "Failed (rmmod)"
exit 1
fi
echo "Done"
;;
status)
/sbin/lsmod | egrep 'Module'
/sbin/lsmod | egrep "${VPNMOD}"
if [ "$?" != "0" ] ; then
echo
echo "Status Failed (lsmod ${VPNMOD}) - The VPN module is not loaded."
fi
echo
/sbin/ifconfig $VPNDEV
if [ "$?" != "0" ] ; then
echo
echo "Status Failed (ifconfig ${VPNDEV}) - The virtual interface is not
present."
exit 1
fi
;;
restart|reload)
$0 stop
$0 start
;;
*)
echo "Usage: $0 {start|stop|restart|reload|status}"
exit 1
esac

exit 0
.....................................................................

I didn't quite understand if you also need the Intel536_boot script from
/etc/init.d. Isn't this the same as from your tarball?

The shell output of your commands are as follows:

linux:~/Installationen/Modem-536ep/intel-536-537 # ls -l
/dev/modem
lrwxrwxrwx 1 root root 10 17. Feb 17:56 /dev/modem -> /dev/536ep
linux:~/Installationen/Modem-536ep/intel-536-537 # ls -l
/dev/536*
crw-rw-rw- 1 root dialout 240, 1 17. Feb 17:56 /dev/536ep
crw------- 1 root root 240, 1 17. Feb 17:56 /dev/536ep0
linux:~/Installationen/Modem-536ep/intel-536-537 #
/etc/init.d/Intel536_boot status
Intel536 1079683 0
hamregistry NOT running
linux:~/Installationen/Modem-536ep/intel-536-537 #
/etc/init.d/Intel536_boot stop
hamregistry: Kein Prozess gefunden

linux:~/Installationen/Modem-536ep/intel-536-537 #
/etc/init.d/Intel536_boot start

tail -f /var/log/messages
Feb 17 20:22:01 linux su: (to root) junker on /dev/pts/0
Feb 17 20:24:54 linux kernel: [ 6462.907952] Intel 536EP card
found
Feb 17 20:24:54 linux modem-manager: (tty/536ep0): could not get
port's parent device

I'll try next the efax and Jacques' hint concerning the device-link
(thank you for that!).

All the best
Wolfgang

On Thu, 16 Feb 2012, Philippe Vouters wrote:

Dear Wolfgang,

First it does look like that the intel-536-537/Intel536_inst shell
code we have been working on needs some improvements for a SUSE
distribution to avoid the warning messages.

To improve this code, can you send me one but the Intel536_boot and
vpnclient_init scripts. Both of these two are wrong for a SUSE
distribution. You will find one in /etc/init.d/. I shall use it as a
model for correct SUSE boot script programming.

However now, the Intel536.ko driver gets correctly installed in
/lib/modules/2.6.37.1-1.2-default/kernel/drivers/char/Intel536.ko

Can you from two terminals do the following ?

From one user account terminal:
$ tail -f /var/log/messages

From a root account terminal:
ls -l /dev/modem
ls -l /dev/536*
/etc/init.d/Intel536_boot status
/etc/init.d/Intel536_boot stop
/etc/init.d/Intel536_boot start

And you send me both screen captures.

There is one message which shows something really wrong that I cannot
figure out where the fault comes from:
Feb 16 18:12:33 linux modem-manager: (tty/536ep0): could not get port's
parent device
Why tty in tty/536ep0 ???? And this message comes from linux
modem-manager, a software I have no control over and I never used on
my Fedora distribution.

Anyhow and from now on, reload my
http://vouters.dyndns.org/Intel/Intel-Readme.html Print out its
content and follow carefully each step of the guide. Prior to testing
with wvdial, use efax which is either in your /usr/bin directory or
that you can download from http://vouters.dyndns.org/Intel/ in source
zip format. To build the binary:
$ cd efax
$ make -f Makefile_original
If ls /usr/bin/efax says nothing, then replace efax by ./efax in the
command I document.

Philippe

Le 16/02/2012 19:47, Wolfgang Junker a écrit :
Dear Philippe,

I followed your advice by putting "set -x" into the Intel536_inst
and commenting out the three lines after "D=...". If I do
$make install
I get the following output:

make[1]: Entering directory
`/root/Installationen/Modem-536ep/intel-536-537'
rm -f /etc/hamregistry.bin
bash Intel536_inst
++ uname -r
+ KERNVER=2.6.37.1-1.2-default
+ echo 'running kernel 2.6.37.1-1.2-default'
running kernel 2.6.37.1-1.2-default
+ '[' ']'
+ CharModDir=/lib/modules/2.6.37.1-1.2-default/kernel/drivers/char
+ case $KERNVER in
+ KMS=ko
+ '[' '!' -d /lib/modules/2.6.37.1-1.2-default/kernel/drivers/char ']'
+ '[' -a ./hamregistry.bin ']'
+ rm -f /etc/hamregistry.bin
+ echo 'installing hamregistry, used for persistant storage'
installing hamregistry, used for persistant storage
+ install -o root -g root -m 110 hamregistry /usr/sbin
+ echo 'installing Intel536 driver'
installing Intel536 driver
+ install -o root -g root -m 744 Intel536.ko
/lib/modules/2.6.37.1-1.2-default/kernel/drivers/char/Intel536.ko
++ tr '[:upper:]' '[:lower:]'
++ ls /etc/lsb-release /etc/SuSE-release
+ D='/etc/lsb-release
/etc/suse-release'
+ case $D in
+ DISTRIB_ID=SUSE
+ case $DISTRIB_ID in
++ cut -f3 '-d '
++ grep VERSION /etc/SuSE-release
+ DISTVER=11.4
+ echo 'SuSE version 11.4'
SuSE version 11.4
+ case $DISTVER in
+ echo 'SuSE boot scripts'
SuSE boot scripts
+ INITDIR=/etc/init.d
+ echo 'SuSE rc3.d and rc5.d scripts'
SuSE rc3.d and rc5.d scripts
+ [[ 11.4 < 10. ]]
+ echo '#!/bin/sh >' /etc/init.d/536EP
#!/bin/sh > /etc/init.d/536EP
+ echo '### BEGIN INIT INFO >>' /etc/init.d/536EP
### BEGIN INIT INFO >> /etc/init.d/536EP
+ echo '# Provides: Intel536_boot'
+ echo '# Required-Start: $syslog'
+ echo '# Required-Stop: $syslog'
+ echo '# Default-Start: 3 5'
+ echo '# Default-Stop: 1 2'
+ echo '# Description: Start Intel 536EP driver'
+ echo '### END INIT INFO'
+ cat /etc/init.d/536EP Intel536_boot
+ chown root.root /etc/init.d/Intel536_boot
+ chmod 755 /etc/init.d/Intel536_boot
+ insserv /etc/init.d/Intel536_boot
insserv: warning: script 'K01vpnclient_init' missing LSB tags and
overrides
insserv: warning: script 'Intel536_boot' missing LSB tags and
overrides
insserv: warning: script 'Intel536_boot' missing LSB tags and
overrides
insserv: Default-Start undefined, assuming default start
runlevel(s) for
script `Intel536_boot'
insserv: warning: script 'vpnclient_init' missing LSB tags and
overrides
+ rm /etc/init.d/536EP
+ echo 'starting module and utilities'
starting module and utilities
+ /sbin/depmod -A
+ /bin/bash Intel536_boot restart
hamregistry: Kein Prozess gefunden

+ echo done
done
make[1]: Leaving directory
`/root/Installationen/Modem-536ep/intel-536-537'

The module Intel536.ko is being loaded, the devices /dev/536ep and
/dev/536ep0 are created and the device /dev/modem as a link to
/dev/536ep. The Intel536_boot-script goes into /etc/init.d,
/etc/init.d/rc3.d and /etc/init.d/rc5.d. All seems to be fine, but
when
I try $wvdialconf I still get the answer

Sorry, no modem was detected!

I did a reboot of the machine to look for boot-messages, and in fact
I get

Feb 16 18:12:33 linux kernel: [ 4277.001953] Intel 536EP card found
Feb 16 18:12:33 linux modem-manager: (tty/536ep0): could not get
port's
parent device

Also, in the IRQ-list the modem does not show up. What might be the
reason for this?

Many thanks for your help!
Wolfgang

On Wed, 15 Feb 2012, Philippe Vouters wrote:

Dear Wolfgang,

Outside subject and first, I am very surprised nm outputs a bunch of
thousand symbols as one of the shell instructions which composes
the $
make 536 is to strip the symbols. I surely missed something here.

Back to your concern, an installable Intel536.ko by $ sudo make
install. As far as I understand what is happening on your side, the
first D=`ls /etc/*-release 2>/dev/null | tr [:upper:] [:lower:]`
assignment in intel536-537/Intel536_inst sounds correct. It returns
"/etc/lsb-release /etc/suse-release" with the "suse" substring being
exploitable.

However as it also returns the substring "lsb", the second D=`sed -r
's/DISTRIB_ID=(\w+)/\1/' /etc/lsb-release | tr [:upper:]
[:lower:]` leads to an unexploitable result.

As you are the very first one reporting me this bug despite me
maintaining this Intel code for many years, before I do change
anything in it, I must be aware how all the other handled Linux
distributions return the first D assignment. If none of them
return an
lsb substring, I shall comment out this second D assignment. If
anyone
of them displays lsb, then I must be aware on what is returned by the
second D assignment.

On my Linux Fedora 16 ,here is what tells the first D assignment:
[philippe@victor intel-536-537]$ echo $D
/etc/fedora-release /etc/redhat-release /etc/system-release
So no lsb substring and the redhat substring being exploitable for
the
rest of the code.

As a first and temporary conclusion for us to progress toward your
satisfaction, edit with the editor of your choice the file
intel536-537/Intel536_inst and manage to have the lines:

# determine distribution:
D=`ls /etc/*-release 2>/dev/null | tr [:upper:] [:lower:]`;

case $D in
*lsb*) D=`sed -r 's/DISTRIB_ID=(\w+)/\1/' /etc/lsb-release | tr
[:upper:] [:lower:]`;;
esac

look like:

# determine distribution:
D=`ls /etc/*-release 2>/dev/null | tr [:upper:] [:lower:]`;

# case $D in
# *lsb*) D=`sed -r 's/DISTRIB_ID=(\w+)/\1/' /etc/lsb-release | tr #
[:upper:] [:lower:]`;;
# esac

This way the second "case $D" just after should correctly recognize
your distribution as being a SUSE distribution.

As you are a very trustful collaborator, I suggest you also temporary
add in this file a "set -x" without the double quotes on top of it so
that it looks like:

#***********************************************************************************

set -x
KERNVER=`uname -r`
echo "running kernel $KERNVER"

This way everyone will be able to control how well this shell code
executes.

Best regards,
Philippe

Le 15/02/2012 22:35, Wolfgang Junker a écrit :
Dear Philippe,

I am happy to assist you in solving the problem!

First of all, $nm Intel536.ko outputs a bunch of several thousand
symbols.

For your first D, $ echo $D gives the output
"/etc/lsb-release /etc/suse-release"

and for the second D
lsb_version="core-2.0-noarch:core-3.2-noarch:core-4.0-noarch:core-2.0-ia32:

core-3.2-ia32:core-4.0-ia32"

Sincerely
Wolfgang

On Tue, 14 Feb 2012, Philippe Vouters wrote:

Dear Wolfgang,

I notice you are clever and it's pretty enjoyable to work with
you. I
thought my find command was correct. However I neglected to test it
before submitting it to you and this is my fault.

The $ make 536 warning you read is quite expected and has always
existed since the very beginning I inheird this code from Intel
Corp.
The fact is that Intel did not provide all necessary files for a
clean
make with no warning. Just for your curiosity : if you perform the
command
$ ls -la coredrv/
that there are two *core.lib files with no *core.lib.cmd unlike the
other files when you will build them with make 536 and if you do
the
command:
$ file <on each of these two *core.lib>
you'll see they are 32-bits ELF binary files. This is because of
these
two binary files, I claim Linux 32 bits support only in my
Intel-Readme.html on my Web site.

Now back to something you are expecting : a working correctly
installed 536EP driver which you can use to connect to Internet via
wvdial.

First and to check everything is indeed correct you should now
have an
Intel536.ko file in your intel-536-537/ directory. To even make it
surer, a:
$ nm Intel536.ko
ought to say no symbols

If the above is quite correct, let us figure out why $ sudo make
install fails on your computer returning the string:
unknown distribution - no boot scripts have been installed
The answer should be in intel536-537/Intel536_inst on the shell
code
line you may test outside the burden of the $ make install
First:
$ D=`ls /etc/*-release 2>/dev/null | tr [:upper:] [:lower:]`
$ echo $D
Next if echo $D display lsb
D=`sed -r 's/DISTRIB_ID=(\w+)/\1/' /etc/lsb-release | tr [:upper:]
[:lower:]`
echo $D should contain on you side the string "suse" as you say you
run a SUSE distribution. As it seems and as a fact from your cut'n
paste, echo $D should contain no "suse" string, nor any of the
other
recognized and dealt with distributions:
mandrake, redhat, redflag, conectiva, ubuntu, debian,
slackware,gentoo, knoppix.

So please mail us what tells each echo $D above. With this
information, I shall be able to guide you aiming at your full
satisfaction.

However, more mail exchanges and work need to be performed in close
collaboration with you. Currently I am on a totally dark land with
your Linux distribution. Two items to be known with your actual
distribution : where should the init scripts be installed ? Next
what
is its common method to activate them at boot time ?

With my warmest regards,
Philippe

Le 14/02/2012 22:47, Wolfgang Junker a écrit :
Dear Philippe,

your find-command does not give any output on my computer. I
nevertheless tried your suggestion, to change the occurances of
(2,6,38)
to (2,6,37) in coredrv.c, locks.c and softserial_io.c .
And in fact, the $ make 536 goes through then with only a warning

WARNING: could not find
/root/Installationen/Modem-536ep/intel-536-537/coredrv/.536core.lib.cmd

for
/root/Installationen/Modem-536ep/intel-536-537/coredrv/536core.lib

and the $ make install goes through with the warning

unknown distribution - no boot scripts have been installed

However, if I follow your advices in the readme.txt, to make the
modem work

$ insmod -f Intel536.ko
$ mknod /dev/536ep c 240 1
$ ln -s /dev/536ep /dev/modem

then no modem is detected, e.g. $ wvdialconf replies

Sorry, no modem was detected!

or if I start kppp and scan for the modem, then the computer
freezes. So
at the moment I do not understand what exactly happens.

Sincerely,
Wolfgang

On Mon, 13 Feb 2012, Philippe Vouters wrote:

Wolfgang,

In coredrv/coredrv.c which fails to compile at line 779 on your
side,
I do the following:
#if LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,38)
#define DECLARE_MUTEX DEFINE_SEMAPHORE
#endif

DECLARE_MUTEX(exec_reg_sem);
DECLARE_WAIT_QUEUE_HEAD(persistWriteQ);
DECLARE_WAIT_QUEUE_HEAD(persistReadQ);
DECLARE_WAIT_QUEUE_HEAD(persistShutdownQ);

The above means that if kernel is 2.6.38 or higher, the
DECLARE_MUTEX(exec_reg_sem) reads as
DEFINE_SEMAPHORE(exec_reg_sem)
I might perhaps have missed that kernel 2.6.37 (your kernel
version)
also needs DEFINE_SEMAPHORE.

To make sure, may I ask you to do the following on your computer:
$ find /lib/modules/`uname -r`/build/include \
-exec grep DEFINE_SEMAPHORE {} \; -print

At fist glance, this find command should return matches on your
system.

If I refer to your experience, all
LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,38) must be changed to
LINUX_VERSION_CODE >= KERNEL_VERSION(2,6,37)
The changes to be brought are in files coredrv/coredrv.c,
coredrv/locks.c and coredrv/softserial_io.c

If the find command displays DEFINE_SEMAPHORE, you may change the
code
yourself. You just edit those three files searching for 2,6,38
and
replace every occurence by 2,6,37. Then you tell us whether a $
make
536
now builds just as fine under your kernel version.

Thank you in advance to keep us informed.

Philippe

Le 13/02/2012 19:56, Wolfgang Junker a écrit :
Dear Philippe,

it's only now that I come back to my computer. Here are the
results of
your shell-commands:

$ uname -r|grep "2.6" --> 2.6.37.1-1.2-default
$ make INTEL_MODEM=536EP 536core_26

make -C /lib/modules/2.6.37.1-1.2-default/build
M=/root/Installationen/Modem-536ep/intel-536-537/coredrv modules
make[1]: Entering directory
`/usr/src/linux-2.6.37.1-1.2-obj/i386/default'
make -C ../../../linux-2.6.37.1-1.2
O=/usr/src/linux-2.6.37.1-1.2-obj/i386/default/. modules
CC [M]
/root/Installationen/Modem-536ep/intel-536-537/coredrv/coredrv.o
/root/Installationen/Modem-536ep/intel-536-537/coredrv/coredrv.c:779:4:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/coredrv.c:779:4:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/coredrv.c:779:4:

warning: parameter names (without types) in function declaration
CC [M]
/root/Installationen/Modem-536ep/intel-536-537/coredrv/clmmain.o
CC [M]
/root/Installationen/Modem-536ep/intel-536-537/coredrv/rts.o
CC [M]
/root/Installationen/Modem-536ep/intel-536-537/coredrv/task.o
CC [M]
/root/Installationen/Modem-536ep/intel-536-537/coredrv/uart.o
CC [M]
/root/Installationen/Modem-536ep/intel-536-537/coredrv/wwh_dflt.o

CC [M]
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.o
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:176:1:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:176:1:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:176:1:

warning: parameter names (without types) in function declaration
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:177:1:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:177:1:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:177:1:

warning: parameter names (without types) in function declaration
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:178:1:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:178:1:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:178:1:

warning: parameter names (without types) in function declaration
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:179:1:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:179:1:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:179:1:

warning: parameter names (without types) in function declaration
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:180:1:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:180:1:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:180:1:

warning: parameter names (without types) in function declaration
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:181:1:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:181:1:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:181:1:

warning: parameter names (without types) in function declaration
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:182:1:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:182:1:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:182:1:

warning: parameter names (without types) in function declaration
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:183:1:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:183:1:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:183:1:

warning: parameter names (without types) in function declaration
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:184:1:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:184:1:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:184:1:

warning: parameter names (without types) in function declaration
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:
In
function ‘linux_RSAGetExclusive’:
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:192:24:

error: ‘sem0’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:192:24:

note: each undeclared identifier is reported only once for each
function
it appears in
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:196:24:

error: ‘sem1’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:200:24:

error: ‘sem2’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:204:24:

error: ‘sem3’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:208:24:

error: ‘sem4’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:212:24:

error: ‘sem5’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:216:24:

error: ‘sem6’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:220:24:

error: ‘sem7’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:224:24:

error: ‘sem8’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:
In
function ‘linux_RSAFreeExclusive’:
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:242:11:

error: ‘sem0’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:245:11:

error: ‘sem1’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:248:11:

error: ‘sem2’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:251:11:

error: ‘sem3’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:254:11:

error: ‘sem4’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:257:11:

error: ‘sem5’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:260:11:

error: ‘sem6’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:263:11:

error: ‘sem7’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:266:11:

error: ‘sem8’ undeclared (first use in this function)
make[4]: ***
[/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.o]
Fehler 1
make[3]: ***
[_module_/root/Installationen/Modem-536ep/intel-536-537/coredrv]
Fehler 2
make[2]: *** [sub-make] Fehler 2
make[1]: *** [all] Fehler 2
make[1]: Leaving directory
`/usr/src/linux-2.6.37.1-1.2-obj/i386/default'
make: *** [536core_26] Fehler 2

cp Intel536.ko .. --> file couldn't be found

Wolfgang

On Sun, 12 Feb 2012, Philippe Vouters wrote:

Dear Wolfgang,

I do not know what can be wrong.
Try the following:
$ cd intel-536-537
$ uname -r|grep "2.6" && \ cd coredrv && make INTEL_MODEM=536EP
536core_26 && \
cp Intel536.ko .. && cd .. && \
strip --strip-debug Intel536.ko

If the above does not output CC and LD strings, try
$ cd coredrv
$ make INTEL_MODEM=536EP 536core_26

and reply to everyone with the cut and paste of what your
read on
your
terminal screen.

Thanks in advance.

Philippe

Le 12/02/2012 18:31, Wolfgang Junker a écrit :
Philippe,

I may be misinterpreting your words, but you write

"If you run a 2.6.x kernel, then you will need
../Intel/intel-536EP-537EP_2011_12_03.tar.bz2"

Since I am running the kernel 2.6.37.1-1.2-default of SUSE
11.4 I
downloaded the intel-536EP-537EP_2011_12_03.tar.bz2, but
obtain
after
"make 536" the following output, reported in my previous
message:

Module precompile check
Current running kernel is: 2.6.37.1-1.2-default
/lib/modules... autoconf.h exists
diff: /boot/vmlinuz.autoconf.h: Datei oder Verzeichnis nicht
gefunden
autoconf.h matches running kernel
diff: /boot/vmlinuz.version.h: Datei oder Verzeichnis nicht
gefunden
version.h matches running kernel
make[1]: Entering directory
`/root/Installationen/Modem-536ep/intel-536-537'
2.6.37.1-1.2-default
Failed to build driver

(the german text in lines 4 and 6 means "couldn't find file or
directory")

Sincerely
Wolfgang

On Sun, 12 Feb 2012, Philippe Vouters wrote:

Wolfgang,

You did not pay enough attention to what I write in the
TARBALLS
KERNEL COMPATIBILITY: chapter of my
http://vouters.dyndns.org/Intel/Intel-Readme.html document.
Download
the correct tarball suited for a 2.6.x Linux kernel which
your
SUSE
11.4 runs as per your indications.
Philippe

Le 12/02/2012 16:41, Antonio Olivares a écrit :
Wolfgang,

This is a new problem that has arised since files were
changed.
Phillipe has made some changes to the code to accomodate
them.
However, they were for newer kernels. Adding CC to Phillipe
so he
can
make some recommendations.

Phillipe,

Wolfgang has Intel-536ep-Modem in his Notebook and want for
it to
run
under SUSE 11.4 with kernel 2.6.37.1-1.2-default. I
recommended
the
latest driver from your site, but he appears to be having
problems.

http://vouters.dyndns.org/Intel/intel-536EP-537EP_2012_01_04.tar.bz2

http://vouters.dyndns.org/Intel/

Thanks for your hard work and improvements to Intel
536/Intel
537
family of modems and other contributions to the community.

Regards,

Antonio

On Sun, Feb 12, 2012 at 9:34 AM, Wolfgang Junker
<junker@xxxxxxxxxxxxxxxxxxx> wrote:
Antonio,

thank you for pointing the way to the newest releases.
Unforntunately, I
have still no success:

If I try the intel-536EP-537EP_2012_01_04.tar.bz2 I get the
same
error log
as described in my previous message.

If I try the intel-536EP-537EP_2011_12_03.tar.bz2 I obtain
the
follwing
output:

Module precompile check
Current running kernel is: 2.6.37.1-1.2-default
/lib/modules... autoconf.h exists
diff: /boot/vmlinuz.autoconf.h: Datei oder Verzeichnis
nicht
gefunden
autoconf.h matches running kernel
diff: /boot/vmlinuz.version.h: Datei oder Verzeichnis nicht
gefunden
version.h matches running kernel
make[1]: Entering directory
`/root/Installationen/Modem-536ep/intel-536-537'
2.6.37.1-1.2-default
Failed to build driver

Wolfgang

On Sat, 11 Feb 2012, Antonio Olivares wrote:

Wolfgang,

Phillipe Vouters has released newer releases, but they
are not
present
in linmodems Intel tree. Check here:

http://vouters.dyndns.org/Intel/intel-536EP-537EP_2012_01_04.tar.bz2

for a newer release. Also you may visit the following
page to
check
for other packages

http://vouters.dyndns.org/Intel/

and Phillipe's great readme.

http://vouters.dyndns.org/Intel/Intel-Readme.html

Should you encounter any difficulties, please let us know.

Regards,

Antonio

2012/2/11 Wolfgang Junker<junker@xxxxxxxxxxxxxxxxxxx>:

Hi all,

I have an Intel-536ep-Modem in my Notebook and want to
run it
under SUSE
11.4 with kernel 2.6.37.1-1.2-default. So I downloaded
the
intel-536EP-537EP_2011_07_03.tar from

http://linmodems.technion.ac.il/packages/intel/Philippe.Vouters

and tried to compile it, but with no success. When I do
"make
536"
I get
the
following error messages:

/root/Installationen/Modem-536ep/intel-536-537/coredrv/coredrv.c:779:4:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/coredrv.c:779:4:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/coredrv.c:779:4:

warning: parameter names (without types) in function
declaration

...

/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:148:1:

warning: data definition has no type or storage class
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:148:1:

warning: type defaults to ‘int’ in declaration of
‘DECLARE_MUTEX’
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:148:1:

warning: parameter names (without types) in function
declaration

...

/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:

In

function
‘linux_RSAGetExclusive’:
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:164:24:

error: ‘sem0’ undeclared (first use in this function)
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:164:24:

note:
each undeclared identifier is reported only once for each
function it
appears in
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:168:24:

error: ‘sem1’ undeclared (first use in this function)

...

/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:

In

function
‘linux_RSAFreeExclusive’:
/root/Installationen/Modem-536ep/intel-536-537/coredrv/locks.c:214:11:

error: ‘sem0’ undeclared (first use in this function)

...

Failed to build driver

Does anyone understand, what is happening and can help
me?
Many thanks in advance

Wolfgang

--
Philippe Vouters (Fontainebleau/France)
URL: http://vouters.dyndns.org/