iDuino - mám je

Jak jsem psal, objednal jsem si 5. srpna od Fundamental Logic 5ks iDuino.



Potvrzení mi přišlo hned, za dva dny pak že se čeká na sestavení. A potom dlouho nic. Na webu sice mají napsáno, že sestavení může trvat až pět pracovních dní, ale 14. srpna už mi to přišlo poněkud moc, tak jsem se připomněl decentním mailem. Druhý den mi odpověděl Kevin Bralten, že jim akorát dorazily od dodavatele a testují je. Zpoždění prý měl na svědomí dodavatel. 21. srpna mi pak dorazily dva maily - jeden od PayPalu a druhý od FL, informující o tom, že balíček veze Canada Post.

Balíček, jedna bublinková obálka v druhé, jsem měl ve schránce přesně tři týdny po objednání - v úterý, ale až teď jsem se dostal k tomu je vyzkoušet - byla to opravdu hračka - z menu vybrat 'Examples-Digital-Blink' na blikání LED, v 'Tools-Serial Port' vybrat /dev/ttyUSB0 a pak jen 'Upload to I/O Board'. Abych měl jistotu, zkusil jsem ještě komunikaci s PC - 'Examples-Communication-ASCIITable' posílá do počítače ASCII tabulku.



Nakonec jsem neplatil žádné clo ani DPH, takže mě jeden kus vyšel na 388,20 Kč. Jeden už je zamluvený, takže 1-2 mohu někomu odprodat.

iDuino

Na svém oblíbeném hackaday.com jsem po vlně různých hacků s platformou Arduino (Vývojová sada s procesorem ATmega168, opensource je hardware i software, soft je multiplatformní) narazil na zprávičku o stickDuino, forku Arduina, který lze zasunout přímo do USB portu. To mě, spolu s informací, že to prodávají za 20 USD, navnadilo. Dolar je dneska za ~15 Kč a to už vychází o něco líp než 770 Kč za echt Arduino Diecimila v českém shopu. Nakonec jsem se ale rozhodl pro iDuino.





iDuino se k PC připojuje pomocí USB s Mini-B konektorem (to je ten malej k digitálním foťákům apod.). Jeho velkou výhodou je maximální kombinovatelnost s nepájivými kontaktními poli.



V procesoru, který je na desce, zabírá 2kB z 16kB bootloader, díky kterému je nahrávání programů jednoduché. Ale pomocí jednoduchých triků se dá deska použít i k programování čistých procesorů, a to i Mega8, Tiny2313 a Tiny13.



Minimální objednávka už hotové desky byla 5ks. 2-3 kusy určitě využiju, ale zbylé dva klidně prodám za náklady (cca 390 Kč, ale asi k tomu přijde i DPH a clo, nechám se překvapit). Takže pokud máte někdo zájem, tak se ozvěte.



Jo a taky díky tomu jsem si dneska konečně aktivoval účet na PayPalu.
Nejdřív jsem musel zavolat na linku Servis24 spořky a nastavit "limity pro transakce bez fyzické přítomnosti karty" (přes web to nejde a hodinu na pobočce se mi trávit nechtělo). Nejdřív mi slečna řekla, že to bude aktivovaný až ve středu, že jim to tak prostě v systému chodí, ale když jsem vyjádřil rozčarování, tak z ní vylezlo, že za 50 Kč to můžu mít expres za 30 minut. No tak jo. Asi za hodinu jsem to zkusil, v internetbankingu sice ještě změna nebyla, ale PayPal už mi kartu ověřil.

Customizing Asus AM200g - V. firmware modification

I would not dare to upload a firmware built from scratch, but what about only slightly modified one?

For me, it would suffice to just add a call to some script on the flash drive that would be launched at boot. From there I'd be able to do all I need (ATM).

The WL-600g GPL Makefile contains the following default sequence of commands in its buildimage section:

./buildFS
$(HOSTTOOLS_DIR)/mksquashfs $(TARGET_FS) $(PROFILE_DIR)/rootfs.img -noappend -be -lzma -no-fragments -noI
$(OBJCOPY) -O binary vmlinux vmlinux.bin
$(HOSTTOOLS_DIR)/cmplzma -k -2 vmlinux vmlinux.bin vmlinux.lz;\
$(HOSTTOOLS_DIR)/bcmImageBuilder --output $(CFE_FS_KERNEL_IMAGE_NAME) --chip $(BRCM_CHIP) --board $(BRCM_BOARD_ID) \
--productname $(CUSTOMER_PRODUCT_NAME) --blocksize $(BRCM_FLASHBLK_SIZE) --cfefile $(CFE_FILE) --rootfsfile rootfs.img --kernelfile vmlinux.lz --include-cfe;

Let's keep hoping that WL-600g is the same as AM-200g, try to reverse the process, do the modifications and put together again.
bcmImageBuilder should just put together the filesystem file, kernel, cfe file (st like bios - see OpenWRT - Everything you need to know about broadcom hardware (Part 1)) and some checksums. There is a utility done by OpenWRT people that should be a replacement of bcmImageBuilder. Here. From there I learned the structure of the header, let's repeat it here for the sake of completeness:
The header takes the first 256 bytes

/* Image component */
struct imagecomp {
uint8_t                 address[12];    /* Address of this component as ASCII */
uint8_t                 len[10];        /* Length of this component as ASCII */
};
/* Image tag */
struct imagetag {
uint8_t                 tagver[4];      /*   0 -   3: Version of the tag as ASCII (2) */
uint8_t                 sig1[20];       /*   4 -  23: BCM_MAGIC_1 */
uint8_t                 sig2[14];       /*  24 -  37: BCM_MAGIC_2 */
uint8_t                 chipid[6];      /*  38 -  43: Chip id as ASCII (6345) */
uint8_t                 boardid[16];    /*  44 -  59: Board id as ASCII (96345GW2, etc...) */
uint8_t                 bigendian[2];   /*  60 -  61: "1" for big endian, "0" for little endian */
uint8_t                 imagelen[10];   /*  62 -  71: The length of all data that follows */
struct imagecomp        cfe;            /*  72 -  93: The offset and length of CFE */
struct imagecomp        rootfs;         /*  94 - 115: The offset and length of the root file system */
struct imagecomp        kernel;         /* 116 - 137: The offset and length of the kernel */
uint8_t                 dualimage[2];   /* 138 - 139: use "0" here */
uint8_t                 inactive[2];    /* 140 - 141: use "0" here */
uint8_t                 reserved1[74];  /* 142 - 215: reserved */
uint32_t                imagecrc;       /* 216 - 219: crc of the images (net byte order) */
uint8_t                 reserved2[16];  /* 220 - 235: reserved */
uint32_t                headercrc;      /* 236 - 239: crc starting from sig1 until headercrc (net byte order) */
uint8_t                 reserved3[16];  /* 240 - 255: reserved */
};

Let's decompose it for the firmware from Joyce:

tagver="36 00 00 00" = "6..."
sig1="42 72 6f 61 64 63 6f 6d 20 43 6f 72 70 6f 72 61 74 69 6f 00" = "Broadcom Corporatio."
sig2="76 65 72 2e 20 32 2e 30 00 00 00 00 00 00" = "ver. 2.0......"
chipid="36 33 34 38 00 00" = "6348.."
boardid="39 36 33 34 38 47 57 2d 31 31 00 00 00 00 00 00" = "96348GW-11......"
bigendian="31 00" = "1"
imagelen="33 38 32 32 33 35 36 00 00 00" = "3822356..." #indeed 3822356 + 256 = 3822612 (the size of the firmware file)
cfe="33 32 31 37 30 33 31 31 36 38 00 00 36 32 31 37 32 00 00 00 00 00" = "32170311..62172....." # this starts to be tricky ...
## 62172 is clearly the cfe size, but the cfe itself is slightly different from the files provided in the WL-600g GPL package.
rootfs="33 32 68 31 37 30 39 36 39 36 30 00 00 33 31 33 37 35 33 36 00 00 00" = "3217096960..3137536..." #fs size = 3137536
kernel="33 32 32 30 32 33 34 34 39 36 00 00 36 32 32 36 34 38 00 00 00 00" = "3220234496..622648...." # kernel size = 622648
## 3137536+622648+62172+256=3822612 - all fits so far!
dualimage="00 00"
inactive="00 00"
reserved="00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 00 00 00 00"
imagecrc="c9 ea a8 50"
reserved2=""

The reserved2 part is different from WL-600g - after some experimentation, I found out that the first four bytes are CRC of the fs image and the second four bytes are CRC of the kernel, but was not able to figure out the CRC algorithm.

to extract cfe:
dd bs=1 if=JOYCE_IAD_U2_B_306063520_cfe_fs_kernel of=cfe skip=256 count=62172
to extract fs:
dd bs=1 if=JOYCE_IAD_U2_B_306063520_cfe_fs_kernel of=fs skip=62428 count=3137536
to extract kernel:
dd bs=1 if=JOYCE_IAD_U2_B_306063520_cfe_fs_kernel of=kernel skip=3199964 count=622648

By calling the Asus' bcmImageBuilder command on the extracted files, I got a firmware identical (verified with md5sum) to the original:

../../Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/hostTools/bcmImageBuilder --output bcmimage --chip 6348 --board 96348GW-11 --blocksize 64 --cfefile cfe --rootfsfile fs --kernelfile kernel --include-cfe

So, let's proceed further...

$ file fs
fs: Squashfs filesystem, big endian, version 2.0, 3136535 bytes, 423 inodes, blocksize: 65536 bytes, created: Mon May 28 08:35:59 2007

the file command confirms the fs file is indeed a squashfs image. Let's extract it:

$ unsquashfs -s fs
Reading a different endian SQUASHFS filesystem on fs
Can't find a SQUASHFS superblock on fs

This is already reported in Ubuntu, I reported it to the Mandriva guys as well.
The hand-compiled squashfs 3.3 seemed to work just fine, but it was not able to deal with the Broadcom specific lzma compression. I tried to mount the fs file on the router, but unfortunately mount was not compiled with support for mounting loop devices

# mount -o ro -o loop /var/usb/usb_2/fs /var/usb/a
mount: Mounting /var/usb/usb_2/fs on /var/usb/a failed: Block device required

After some retries I found out about the excellent OpenBox4 people hacking another Broadcom based ADSL router. They developed nb4-unsquash - a tool to extract the content of Broadcom-lzma-compressed squashfs.


But I still did not trust it entirely. I found out I can mount the internal flash and that's enough for me.
First I made a backup of the original fs:

# mkdir /var/usb/a
# mount -o ro /dev/mtdblock0 /var/usb/a
# /var/usb/usb_1/bin/tar cvf /var/usb/usb_2/firmwarefs.tar /var/usb/a/

Copied the filesystem to my PC:

on-my-PC $ mkdir /tmp/fs
on-my-PC $ netcat -l -p 9000 | dd of=/tmp/fs/fs.tar
on-router# /var/usb/usb_1/bin/dd if=/var/usb/usb_2/firmwarefs.tar |/var/usb/usb_1/bin/mipsel-linux-netcat 192.168.1.64 9000

Unpacked (as root because the contents are special files):

$ su
# cd /tmp/fs
# tar xvf fs.tar

Before doing any modifications I tried recreating the squashfs image:

# /home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/hostTools/mksquashfs /tmp/fs/var/usb/a rootfs.img -noappend -be -lzma -no-fragments -noI

and verified with the nb4-unsquash that it looks similar to the original one:

# nb4-unsquash -d new rootfs.img
# nb4-unsquash -d old /tmp/joycefs
# ls -laR old/ > /tmp/oldfslist
# ls -laR new/ > /tmp/newfslist
# kdiff3 /tmp/oldfslist /tmp/newfslist

I saw that all except some modification times was identical, so I put the recreated fs together with the old kernel and cfe:

# /home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/hostTools/bcmImageBuilder --output bcmimage --chip 6348 --board 96348GW-11 --blocksize 64 --cfefile cfe --rootfsfile rootfs.img --kernelfile kernel --include-cfe

Then I took a deep breath and flashed the router with the recreated firmware ... ... ... ... ... ...
And it worked!

Unfortunately it was not possible to simply find a place in the boot sequence where everything is already done and flash is mounted. The /etc/profile, launched by the line ::respawn:-/bin/sh of /etc/inittab, calls /bin/cfm, which takes care of all the ADSL settings and mounting the flash - but it stays loaded and the script does not go on. So after several retries I found a solution - added another line to the /etc/inittab:

$ diff -u inittab.old inittab
--- inittab.old 2008-07-03 00:31:44.000000000 +0200
+++ inittab     2008-07-03 00:31:52.000000000 +0200
@@ -1,5 +1,6 @@
+::respawn:-/bin/myscript.sh
tty2::askfirst:-/bin/sh

that will launch a script that waits until the flash drive is mounted and then passes the command to script residing on the flash:

$ cat bin/myscript.sh
#!/bin/sh
while [ "`echo /var/usb/usb_*`" = "/var/usb/usb_*" ]; do
ping 127.0.0.1 # the shell does not have the sleep builtin, hence using ping as a substitute
done
ping -c 20 127.0.0.1 # wait 20 seconds after the usb is mounted just to be sure load is not too high
/bin/sh /var/usb/usb_2/myscript.sh > /var/usb/usb_2/myscript.log 2>&1

Then I created the new image:

# /home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/hostTools/mksquashfs /tmp/fs/var/usb/a rootfs.img -noappend -be -lzma -no-fragments -noI

Combined with the rest of the lot:

# /home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/hostTools/bcmImageBuilder --output bcmimage --chip 6348 --board 96348GW-11 --blocksize 64 --cfefile cfe --rootfsfile rootfs.img --kernelfile kernel --include-cfe

and flashed. Works like a charm. I put the firmware to the OpenWRT page on AM-200g. direct link



In the next part I'll try to fix the dhcp.



FYI, this is the content of the /var/usb/usb_2/myscript.sh file, launching the jabber daemon:

#!/bin/sh
/bin/iptables -I INPUT -p tcp -i ppp_8_48_1 --dport 5222 -m state --state NEW -j ACCEPT
/bin/iptables -I INPUT -p tcp -i ppp_8_48_1 --dport 5223 -m state --state NEW -j ACCEPT
/bin/iptables -I INPUT -p tcp -i ppp_8_48_1 --dport 5269 -m state --state NEW -j ACCEPT
/var/usb/usb_1/bin/jabberd -c /var/usb/usb_2/jabberd/jabber.xml

FYI, the steps to cross-compile netcat and coreutils (for the tar and dd commands):

$ wget ftp://heanet.dl.sourceforge.net/n/ne/netcat/netcat-0.7.1.tar.bz2
$ tar xvf netcat-0.7.1.tar.bz2
$ cd netcat-0.7.1
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1
$ make
$ make install
$ cd ..
$ wget http://ftp.gnu.org/gnu/coreutils/coreutils-6.9.tar.bz2
$ tar xvf coreutils-6.9.tar.bz2
$ cd coreutils-6.9
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1
$ make
$ make install

Customizing Asus AM200g - IV. jabberd configuration

Customizing Asus AM200g - III. jabberd and vim

Customizing Asus AM200g - II. Preparing the environment

Customizing Asus AM200g - I. PREPARATION

Customizing Asus AM200g - IV. jabberd configuration

The following is a sequence of trial and errors, one by one, covering the path from the default config file to a working one. It is intended to be a search engine trap for people trying to resolve configuration problems.

$ telnet 192.168.1.1
Trying 192.168.1.1...
Connected to r (192.168.1.1).
Escape character is '^]'.
BCM96348 ADSL Router
Login: admin
Password:
> sh
BusyBox v1.00 (2005.04.12-18:11+0000) Built-in shell (msh)
Enter 'help' for a list of built-in commands.
# /var/usb/usb_1/bin/jabberd
Not writing pidfile /var/usb/usb_1/var/run/jabberd/jabber.pid: Read-only file system

In my setup, the jabberd ecosystem resides on a readonly filesystem, so the very first thing to do is to copy the etc/jabber.xml file to the vfat partition and make sure the files that need to be dynamically created/modified are there as well.

# mkdir /var/usb/usb_2/jabberd
# cp /var/usb/usb_1/etc/jabber.xml /var/usb/usb_2/jabberd
# /var/usb/usb_1/bin/vim /var/usb/usb_2/jabberd/jabber.xml

If you're not familiar with vi, here's quick help:

To quit without saving type ':q' (colon q Enter)

To save changes and quit type ':wq' (colon w q Enter)

To move around use the arrow keys

To start writing at cursor press 'i'

To get out of the writing mode (so that you can move somewhere else) press 'Esc'

To delete character at cursor press 'x'




The changes I did at first were:


1.

To prevent the message

Not writing pidfile /var/usb/usb_1/var/run/jabberd/jabber.pid: Read-only file system

I changed /var/usb/usb_1/var/run/jabberd/jabber.pid to /var/usb/usb_2/jabberd/jabber.pid.

2.


Next I launched jabberd in debug mode (-D) and pointed it to the alternate config file (-c path/to/file):

# /var/usb/usb_1/bin/jabberd -D -c /var/usb/usb_2/jabberd/jabber.xml
Mon Jan  3 20:31:13 2000  mio_tls.cc:921 MIO TLS init (GNU TLS)
Mon Jan  3 20:31:14 2000  log.cc:221 <log xmlns='jabber:server' type='warn' from='-internal'>
Cannot open /var/usb/usb_1/etc/dhparams.pem for reading dhparams: No such file or directory</log>
...

The comments in the file say:

<!-- With the <dhparams/> element right inside the <tls/> element,  -->
<!-- you can configure a file containing parameters for Diffie      -->
<!-- Hellmann key exchanges. If this configuration setting is not   -->
<!-- present, jabberd14 will generated these parameters             -->
<!-- automatically on each startup. This takes some time, therefore -->
<!-- you get a faster startup, if this setting is present.          -->

So I commented out <dhparams type='pem'>/var/usb/usb_1/etc/dhparams.pem</dhparams> as I don't know how to make it properly and don't mind waiting a few seconds more at startup. To start a comment, you write "<!--", you close it with "-->".

3.

# /var/usb/usb_1/bin/jabberd -D -c /var/usb/usb_2/jabberd/jabber.xml
...
Mon Jan  3 20:47:02 2000  xdb_sql.cc:911 using the following query on SQL connection establishment: SET NAMES utf8
Mon Jan  3 20:47:02 2000  log.cc:221 <log xmlns='jabber:server' type='alert' from='xdbsql.localhost'>Your xdb_sql is compiled without support
for the selected database driver 'mysql'.</log>
Mon Jan  3 20:47:02 2000  xdb_sql.cc:813 processing handler definition: <handler xmlns='jabber:config:xdb_sql' ns='jabber:iq:last'>
...
Mon Jan  3 20:47:03 2000  deliver.cc:837 delivering to instance 'elogger.localhost'
Mon Jan  3 20:47:03 2000  deliver.cc:614 DELIVER 3:xdbsql.localhost <log xmlns='jabber:server' type='alert' from='xdbsql.localhost'>Your
xdb_sql is compiled without support for the selected database driver 'mysql'.</log>

Yes, this is true, the daemon was compiled without any sql support. I've followed the steps in README.filespool to setup jabber to store data in files - I definitely don't want to run sql server on the machine. Of course I replaced the <xdb_file>$PREFIX/lib/libjabberdxdbfile.so</xdb_file> with <xdb_file>/var/usb/usb_1/lib/libjabberdxdbfile.so</xdb_file>, and <spool><jabberd:cmdline flag='s'>$PREFIX/var/spool/jabberd</jabberd:cmdline></spool> with <spool><jabberd:cmdline flag='s'>/var/usb/usb_2/jabberd/jabberdspool</jabberd:cmdline></spool> - then made sure the directory exists:

# mkdir /var/usb/usb_2/jabberd/jabberdspool

4.

# /var/usb/usb_1/bin/jabberd -D -c /var/usb/usb_2/jabberd/jabber.xml
...
Mon Jan  3 21:17:58 2000  log.cc:221 <log xmlns='jabber:server' type='alert' from='inject.localhost'>
could not open directory /var/usb/usb_1/var/spool/jabberd/inject.localhost for reading</log>
...

inject.localhost is a directory, where some other process can put messages. jabberd then automatically sends them. To make it working I changed
<in>/var/usb/usb_1/var/spool/jabberd/inject.localhost</in> to <in>/var/usb/usb_2/jabberd/inject.localhost</in> and created the directory with

# mkdir /var/usb/usb_2/jabberd/inject

5.

# /var/usb/usb_1/bin/jabberd -D -c /var/usb/usb_2/jabberd/jabber.xml
...
Mon Jan  3 21:33:15 2000  jabberd.cc:336 main load check of 1.00 with 4 total threads
Mon Jan  3 21:33:16 2000  users.cc:209 0        total users
Mon Jan  3 21:33:17 2000  dialback.cc:1207 dialback idle check

hmm, seems to be working ..., time to launch kopete and try registering a user hajma@192.168.1.1. It failed, of course. The message I got from Kopete was definitely not helpful: "There was an error in the protocol stream: There was a conflict in the information received." and "Unable to create account on the server. The Jabber ID is probably already in use.". What a nonsense ... a brief look into the konsole shows the infamous "bouncing a routed packet" message:

Mon Jan  3 21:35:40 2000  deliver.cc:614 DELIVER 4:192.168.1.1 <route xmlns='jabber:server' type='auth'
to='hajma@192.168.1.1' from='14@c2s/100E0278'><iq type='set' to='192.168.1.1' id='aab0a'>
<query xmlns='jabber:iq:register'>
<username>hajma</username>
<password>qwertyuiop</password>
</query>
</iq></route>
Mon Jan  3 21:35:40 2000  deliver.cc:837 delivering to instance 'dnsrv.localhost'
Mon Jan  3 21:35:40 2000  deliver.cc:766 delivery failed (Internal Delivery Error)
Mon Jan  3 21:35:40 2000  log.cc:221 <log xmlns='jabber:server' type='notice' from='192.168.1.1'>bouncing a routed
packet to hajma@192.168.1.1 from 14@c2s/100E0278: Internal Delivery Error</log>

In short, to make jabberd working correctly, you have to replace all occurences of 'localhost' by the domain or IP jabberd is listening at. Now I replaced localhost with 192.168.1.1.

6.

# /var/usb/usb_1/bin/jabberd -D -c /var/usb/usb_2/jabberd/jabber.xml
...

Registering with Kopete again ... This time I only got one error: "Unable to create account on the server. The Jabber ID is probably already in use.". And again it was completely unrelated - in the konsole I found this:

Mon Jan  3 21:50:02 2000  mod_register.cc:278 we got a reply for: username
Mon Jan  3 21:50:02 2000  mod_register.cc:278 we got a reply for: password
Mon Jan  3 21:50:03 2000  mod_register.cc:298 returned err msg: Missing data field: name
...
<error code='400' type='modify'><bad-request xmlns='urn:ietf:params:xml:ns:xmpp-stanzas'/>
<text xmlns='urn:ietf:params:xml:ns:xmpp-stanzas'>Missing data field: name</text></error></iq></route>

The stupid thing Kopete completely ignores XEP-0077, does not ask the server what is needed for registration and just sends username and password. To support Kopete, I had to delete
<name/> and <email/> from the <register xmlns="jabber:iq:register" notify="yes"> section.


After this change I am able to register a new user and send messages.
Just to be perfect I also changed /var/usb/usb_1/var/log/jabberd/error.log to /var/usb/usb_2/jabberd/error.log.
Off to the next level!

7.


To have it available from outside world, I replaced 192.168.1.1 with my_domain.cz.
To open a port in the AM-200g router, issue the following command:

iptables -I INPUT -p tcp -i ppp_8_48_1 --dport 5222 -m state --state NEW -j ACCEPT

8.

The main reason for running own jabber server is more privacy and that is impossible without encryption - let's do it.
To be able to use TLS I had to generate a self-signed certificate. I've used the procedure described at http://www.vanemery.com/Linux/Jabber/jabberd.html:

$ openssl req -new -x509 -newkey rsa:1024 -days 3650 -keyout privkey.pem -out key.pem

Generating a 1024 bit RSA private key

...........++++++

.........++++++

writing new private key to 'privkey.pem'

Enter PEM pass phrase:

Verifying - Enter PEM pass phrase:

Customizing Asus AM200g - III. jabberd and vim

It took me some time to actually make the jabber server working, few people seem to really care about cross-compiling their stuff, but I succeeded and here's the howto:




0. Errata


First of all, during the testing I found out I have to modify the environment settings - the static build of jabber did not work at all. I also added the location of newly compiled libs and headers so that they can be used by programs that depend on them. Some programs were in C++, so the CXX variable had to be set as well. Here's the final environment (P.S. in the end it seems some additions like CPPFLAGS would be needed for perfection, but I'm not in the mood of retrying):

$ cat environment
TOOLCHAINS_HOME=/opt/toolchains/uclibc-crosstools
export LIBS="-L${TOOLCHAINS_HOME}/lib/gcc/mips-linux/3.4.2 \
-L/var/usb/usb_1/lib \
"
export INCLUDES="-I${TOOLCHAINS_HOME}/include/c++/3.4.2 \
-I${TOOLCHAINS_HOME}/mips-linux/sys-include \
-I/var/usb/usb_1/include \
-I/home/hajma/Documents/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/kernel/linux/include \
-I/var/usb/usb_1/include/ncurses \
"
#export CFLAGS="${INCLUDES} ${LIBS} -static"
#export LDFLAGS="-static"
export CFLAGS="${INCLUDES} ${LIBS}"
export LDFLAGS=""
export CC="${TOOLCHAINS_HOME}/bin/mipsel-uclibc-gcc"
export CXX="${TOOLCHAINS_HOME}/bin/mipsel-uclibc-g++"
export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:${TOOLCHAINS_HOME}/lib
export PATH="${TOOLCHAINS_HOME}/bin:${PATH}"

Let's use it:

. environment

Also the symlinking of the uclibc toolchain needed to be further enhanced for some package to get built:

# cd /opt/toolchains/uclibc-crosstools/bin
# for A in mips-linux-uclibc-* ; do
> ln -s ${A} mipsel-linux-${A#mips-linux-uclibc-}
> done

1a. jabberd

$ cd GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource
$ wget http://download.jabberd.org/jabberd14/jabberd14-1.6.1.1.tar.gz
$ tar xvf jabberd14-1.6.1.1.tar.gz
$ cd jabberd14-1.6.1.1
$./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1

This ends badly:

...
checking for libpopt... configure: error: no

Apparently the library for parsing command line parameters is needed. This was reported a year ago ...


2. popt

$ cd GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource
$ wget http://freshmeat.net/redir/popt/72854/url_tgz/popt-1.14.tar.gz
$ tar xvf popt-1.14.tar.gz
$ cd popt-1.14
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1
...
checking for va_copy() function... configure: error: cannot run test program while cross compiling
See `config.log' for more details.

WTF! There's bug 25 about it that the developers closed, saying "va_copy is no longer needed or used in popt-1.14.". Hell, when it's not needed, why did you keep the check ... :-(

I added a comment to the bug and sucessfully run:

$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 ac_cv_va_copy=no
$ make
$ make install

1b. jabberd

Retry:

$ cd ../jabberd14-1.6.1.1
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1
...
checking for libpopt... configure: error: no

The previous step did not help much ... The configure script does not search for the popt.h in the provided directories - again something reported a year ago. It forced me to rise bug 93.

Retry:

$ CPPFLAGS="-I/var/usb/usb_1/include" ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1
...
checking for GNUTLS... no
No package 'gnutls' found
configure: error: Couldn't find required GnuTLS installation

Fine, let's get GnuTLS working first. Furthermore, ./configure --help reveals that some more libraries are needed by jabberd:

--with-libidn=DIR       Where to find libidn (required)
--with-libpth=DIR       Where to find libpth (required)
--with-expat=DIR        Where to find libexpat (required)

In addition GnuTLS requires some other libraries, libtasn1 and libgcrypt, so let's get them all first.


3. GNU Portable Threads

$ cd ..
$ wget ftp://ftp.gnu.org/gnu/pth/pth-2.0.7.tar.gz
$ tar xvf pth-2.0.7.tar.gz
$ cd pth-2.0.7
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1
$ make
$ make install

4. Expat XML Parser

$ cd ..
$ wget http://downloads.sourceforge.net/expat/expat-2.0.1.tar.gz?modtime=1181083143&big_mirror=0
$ tar xvf expat-2.0.1.tar.gz
$ cd expat-2.0.1
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1
$ make
$ make install

5. GNU IDN Library - Libidn

$ cd ..
$ wget ftp://alpha.gnu.org/pub/gnu/libidn/libidn-1.8.tar.gz
$ tar xvf libidn-1.8.tar.gz
$ cd libidn-1.8
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1
$ make
$ make install

6. Tiny ASN.1 library - Libtasn1

$ cd ..
$ wget http://josefsson.org/gnutls/releases/libtasn1/libtasn1-1.4.tar.gz
$ tar xvf libtasn1-1.4.tar.gz
$ cd libtasn1-1.4
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1
$ make
$ make install

To save me from

checking for gpg-error-config... no
checking for GPG Error - version >= 1.4... no
configure: error: libgpg-error is needed.
See ftp://ftp.gnupg.org/gcrypt/libgpg-error/ .

when configuring libgcrypt, libgpg-error has to be done in advance.


7. libgpg-error

$ cd ..
$ wget ftp://ftp.gnupg.org/gcrypt/libgpg-error/libgpg-error-1.6.tar.bz2
$ tar xvf libgpg-error-1.6.tar.bz2
$ cd libgpg-error-1.6
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1
$ make
$ make install

8. libgcrypt

$ cd ..
$ wget ftp://ftp.gnupg.org/gcrypt/libgcrypt/libgcrypt-1.4.1.tar.gz
$ tar xvf libgcrypt-1.4.1.tar.gz
$ cd libgcrypt-1.4.1

The path to libggpg-error has to be specified explicitly:

$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-gpg-error-prefix=/var/usb/usb_1 --with-pth-prefix=/var/usb/usb_1

But it still does not help!:

$ make
...
/opt/toolchains/uclibc-crosstools/bin/mipsel-uclibc-gcc -I/var/usb/usb_1/include -I/opt/toolchains/uclibc-crosstools/include/c++/3.4.2 -I/opt/toolchains/uclibc-crosstools/mips-linux/sys-include -I/var/usb/usb_1/include -I/home/hajma/Documents/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/kernel/linux/include -I/var/usb/usb_1/include/ncurses -Wall -Wpointer-arith -o .libs/version version.o  -L/opt/toolchains/uclibc-crosstools/lib/gcc/mips-linux/3.4.2 -L/var/usb/usb_1/lib ../src/.libs/libgcrypt.so -Wl,--rpath -Wl,/var/usb/usb_1/lib
/opt/toolchains/uclibc-crosstools_gcc-3.4.2_uclibc-20050502/bin/../lib/gcc/mips-linux-uclibc/3.4.2/../../../../mips-linux-uclibc/bin/ld: warning: libgpg-error.so.0, needed by ../src/.libs/libgcrypt.so, not found (try using -rpath or -rpath-link)
../src/.libs/libgcrypt.so: undefined reference to `gpg_err_code_from_syserror'
../src/.libs/libgcrypt.so: undefined reference to `gpg_err_code_from_errno'
../src/.libs/libgcrypt.so: undefined reference to `gpg_strerror'
../src/.libs/libgcrypt.so: undefined reference to `gpg_strsource'
collect2: ld returned 1 exit status
make[2]: *** [version] Error 1
make[2]: Leaving directory `/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/libgcrypt-1.4.1/tests'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/libgcrypt-1.4.1'
make: *** [all] Error 2
$

I found out the guys at OpenWRT had the same problem and have a patch available.
I wanted to file a bug to the GnuPG bug tracker, but did not receive the confirmation mail when registering...
Actually this was told to the author two years ago and he does not care at all.

So here goes the needed modification:

$ diff -u tests/Makefile.in.old tests/Makefile.in
--- tests/Makefile.in.old       2008-06-12 00:01:05.000000000 +0200
+++ tests/Makefile.in   2008-06-12 00:01:11.000000000 +0200
@@ -313,7 +313,7 @@
# a built header.
AM_CPPFLAGS = -I../src -I$(top_srcdir)/src
AM_CFLAGS = $(GPG_ERROR_CFLAGS)
-LDADD = ../src/libgcrypt.la
+LDADD = ../src/libgcrypt.la @LTLIBOBJS@ @GPG_ERROR_LIBS@
# pkbench uses mmap for no good reason.  Needs to be fixed.  Code for
# this can be found in libksba/tests.

and try again:

$ make clean
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-gpg-error-prefix=/var/usb/usb_1 --with-pth-prefix=/var/usb/usb_1
$ make
$ make install

9. The GNU Transport Layer Security Library

$ cd ..
$ wget ftp://ftp.gnutls.org/pub/gnutls/gnutls-2.2.5.tar.bz2
$ tar xvf gnutls-2.2.5.tar.bz2
$ cd gnutls-2.2.5

To prevent a configure error

checking for libgcrypt-config... no
checking for LIBGCRYPT - version >= 1.2.4... no
configure: error:
***
*** libgcrypt was not found. You may want to get it from
*** ftp://ftp.gnupg.org/gcrypt/libgcrypt/
***

the prefix to libgcrypt has to be specified (and libtasn1 too, to avoid using the built-in one):

$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-libgcrypt-prefix=/var/usb/usb_1 --with-libtasn1-prefix=/var/usb/usb_1

but it still does not work:

$ make
...
/opt/toolchains/uclibc-crosstools/bin/mipsel-uclibc-gcc -std=gnu99 -I/opt/toolchains/uclibc-crosstools/include/c++/3.4.2 -I/opt/toolchains/uclibc-crosstools/mips-linux/sys-include -I/var/usb/usb_1/include -I/home/hajma/Documents/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/kernel/linux/include -I/var/usb/usb_1/include/ncurses -D_REENTRANT -D_THREAD_SAFE -pipe -I/var/usb/usb_1/include -I/opt/toolchains/uclibc-crosstools/include/c++/3.4.2 -I/opt/toolchains/uclibc-crosstools/mips-linux/sys-include -I/var/usb/usb_1/include -I/home/hajma/Documents/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/kernel/linux/include -I/var/usb/usb_1/include/ncurses -D_REENTRANT -D_THREAD_SAFE -o ex-cert-select ex-cert-select.o  -L/opt/toolchains/uclibc-crosstools/lib/gcc/mips-linux/3.4.2 -L/var/usb/usb_1/lib ./.libs/libexamples.a ../../gl/.libs/libgnu.a /var/usb/usb_1/lib/libgcrypt.so /var/usb/usb_1/lib/libgpg-error.so ../../lib/.libs/libgnutls.so ../../libextra/.libs/libgnutls-extra.so  -Wl,--rpath -Wl,/var/usb/usb_1/lib -Wl,--rpath -Wl,/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/gnutls-2.2.5/lib/.libs -Wl,--rpath -Wl,/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/gnutls-2.2.5/libextra/.libs -Wl,--rpath -Wl,/var/usb/usb_1/lib
/opt/toolchains/uclibc-crosstools_gcc-3.4.2_uclibc-20050502/bin/../lib/gcc/mips-linux-uclibc/3.4.2/../../../../mips-linux-uclibc/bin/ld: warning: libtasn1.so.3, needed by ../../lib/.libs/libgnutls.so, not found (try using -rpath or -rpath-link)
../../lib/.libs/libgnutls.so: undefined reference to `asn1_write_value@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_delete_structure@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_get_length_der@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_get_tag_der@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_der_decoding_startEnd@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_der_decoding@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_create_element@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_copy_node@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_read_value@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_number_of_elements@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_array2tree@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_check_version@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_find_node@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_find_structure_from_oid@LIBTASN1_0_3'
../../lib/.libs/libgnutls.so: undefined reference to `asn1_der_coding@LIBTASN1_0_3'
collect2: ld returned 1 exit status
make[4]: *** [ex-cert-select] Error 1
make[4]: Leaving directory `/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/gnutls-2.2.5/doc/examples'
make[3]: *** [all-recursive] Error 1
make[3]: Leaving directory `/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/gnutls-2.2.5/doc'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/gnutls-2.2.5/doc'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/gnutls-2.2.5'
make: *** [all] Error 2

After some experimentation (I told you I am no programmer) I found the following to work:

$ LIBS="-ltasn1 $LIBS" ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-libgcrypt-prefix=/var/usb/usb_1 --with-libtasn1-prefix=/var/usb/usb_1
$ make
$ make install

Also I found that the two official mirrors have both different directory structure, so I was missing the devel branch of GnuTLS. The above works with http://www.gnu.org/software/gnutls/releases/gnutls-2.3.14.tar.bz2 (and breaks with 2.2.5).


1c. jabberd

Let's not try to find more bugs and specify the needed components to the configure script now:

$ cd ../jabberd14-1.6.1.1
$ CPPFLAGS="-I/var/usb/usb_1/include" ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-libidn=/var/usb/usb_1 --with-libpth=/var/usb/usb_1 --with-expat=/var/usb/usb_1

but still bad luck:

checking for GnuTLS... checking for mipsel-linux-pkg-config... no
checking for pkg-config... /usr/bin/pkg-config
configure: WARNING: In the future, Autoconf will not detect cross-tools
whose name does not start with the host triplet.  If you think this
configuration is useful to you, please write to autoconf@gnu.org.
checking pkg-config is at least version 0.9.0... yes
checking for GNUTLS... no
No package 'gnutls' found
configure: error: Couldn't find required GnuTLS installation

And again, reported a year ago.
We can fix that, but the show is still not at the end:

$ PKG_CONFIG_PATH=/var/usb/usb_1/lib/pkgconfig CPPFLAGS="-I/var/usb/usb_1/include" ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-libidn=/var/usb/usb_1 --with-libpth=/var/usb/usb_1 --with-expat=/var/usb/usb_1
...
checking for GnuTLS... checking for mipsel-linux-pkg-config... no
checking for pkg-config... /usr/bin/pkg-config
configure: WARNING: In the future, Autoconf will not detect cross-tools
whose name does not start with the host triplet.  If you think this
configuration is useful to you, please write to autoconf@gnu.org.
checking pkg-config is at least version 0.9.0... yes
checking for GNUTLS... yes
checking for TLS 1.2 support in GnuTLS... yes
checking for GNUTLSEXTRA... yes
checking idna.h usability... yes
checking idna.h presence... yes
checking for idna.h... yes
checking for stringprep_check_version in -lidn... no
configure: error: Couldn't find required libidn installation

Quick look into config.log reveals something familiar:

/opt/toolchains/uclibc-crosstools_gcc-3.4.2_uclibc-20050502/bin/../lib/gcc/mips-linux-uclibc/3.4.2/../../../../mips-linux-uclibc/bin/ld: warning: libtasn1.so.3, needed by /var/usb/usb_1/lib/libgnutls-extra.so, not found (try using -rpath or -rpath-link)
/var/usb/usb_1/lib/libgnutls.so: undefined reference to `asn1_write_value@LIBTASN1_0_3'
/var/usb/usb_1/lib/libgnutls.so: undefined reference to `gcry_sexp_release@GCRYPT_1.2'
...

Let's see if the GnuTLS fix works here as well:

LIBS="-ltasn1 $LIBS" PKG_CONFIG_PATH=/var/usb/usb_1/lib/pkgconfig CPPFLAGS="-I/var/usb/usb_1/include" ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-libidn=/var/usb/usb_1 --with-libpth=/var/usb/usb_1 --with-expat=/var/usb/usb_1
...
checking for stringprep_check_version in -lidn... no

No, it doesn't :-(, but config.log reveals we got one step further:

...
/opt/toolchains/uclibc-crosstools/bin/mipsel-uclibc-g++ -o conftest -g -O2 -I/va
r/usb/usb_1/include -I/var/usb/usb_1/include -I/var/usb/usb_1/include -I/var/usb
/usb_1/include conftest.cpp  -L/var/usb/usb_1/lib /var/usb/usb_1/lib/libgnutls-e
xtra.so /var/usb/usb_1/lib/libgnutls.so /var/usb/usb_1/lib/libidn.so -lresolv /v
ar/usb/usb_1/lib/libtasn1.so -L/opt/toolchains/uclibc-crosstools/lib/gcc/mips-li
nux/3.4.2 -ldl -Wl,--rpath -Wl,/var/usb/usb_1/lib -Wl,--rpath -Wl,/var/usb/usb_1
/lib
/opt/toolchains/uclibc-crosstools_gcc-3.4.2_uclibc-20050502/bin/../lib/gcc/mips-
linux-uclibc/3.4.2/../../../../mips-linux-uclibc/bin/ld: warning: libgcrypt.so.1
1, needed by /var/usb/usb_1/lib/libgnutls-extra.so, not found (try using -rpath
or -rpath-link)
...

Finally, this works:

LIBS="-ltasn1 -lgpg-error -lgcrypt $LIBS" PKG_CONFIG_PATH=/var/usb/usb_1/lib/pkgconfig CPPFLAGS="-I/var/usb/usb_1/include" ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-libidn=/var/usb/usb_1 --with-libpth=/var/usb/usb_1 --with-expat=/var/usb/usb_1

but make fails:

$ make
...
/var/usb/usb_1/include/gnutls/openpgp.h: In function `void mio_tls_process_credentials(xmlnode_t*, const std::list<std::string, std::allocator<std::string> >&, const std::list<std::string, std::allocator<std::string> >&, gnutls_dh_params_int*, const std::list<std::string, std::allocator<std::string> >&, const std::list<std::string, std::allocator<std::string> >&)':
/var/usb/usb_1/include/gnutls/openpgp.h:247: error: too few arguments to function `int gnutls_certificate_set_openpgp_key_file(gnutls_certificate_credentials_st*, const char*, const char*, gnutls_openpgp_crt_fmt_t)'
mio_tls.cc:615: error: at this point in file
/var/usb/usb_1/include/gnutls/openpgp.h:271: error: too few arguments to function `int gnutls_certificate_set_openpgp_keyring_file(gnutls_certificate_credentials_st*, const char*, gnutls_openpgp_crt_fmt_t)'
mio_tls.cc:634: error: at this point in file
mio_tls.cc:653: error: `gnutls_certificate_set_openpgp_trustdb' undeclared (first use this function)
mio_tls.cc:653: error: (Each undeclared identifier is reported only once for each function it appears in.)
make[3]: *** [mio_tls.lo] Error 1
make[3]: Leaving directory `/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/jabberd14-1.6.1.1/jabberd'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/jabberd14-1.6.1.1/jabberd'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/jabberd14-1.6.1.1'
make: *** [all] Error 2

This turned out to be easy, the Gentoo folks found and patched it couple of months ago (and it looks it is in the jabberd cvs now) and after applying it, the journey ends:

$ make
$ make install

Wow, congratulations!!


Let's get an editor so that we can modify the config files in place:


10a. vim

$ cd ..
$ wget ftp://ftp.vim.org/pub/vim/unix/vim-7.1.tar.bz2
$ tar xvf vim-7.1.tar.bz2
$ cd vim71
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-features=tiny
...
checking for strip... strip
configure: error: cannot compile a simple program, check CC and CFLAGS
(cross compiling doesn't work)

Would you expect an opensource app as big and widespread as Vim to be so broken? I would now. Fortunately the OpenWRT and FreeWRT people come to help again.

$ wget --no-check-certificate https://dev.openwrt.org/browser/packages/utils/vim/patches/001-cross_compile.patch?format=txt
$ patch -p1 <001-cross_compile.patch\?format\=txt
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-features=tiny
...
checking --with-tlib argument... empty: automatic terminal library selection
checking for tgetent in -lncurses... no
checking for tgetent in -ltermlib... no
checking for tgetent in -ltermcap... no
checking for tgetent in -lcurses... no
no terminal library found
checking for tgetent()... configure: error: NOT FOUND!
You need to install a terminal library; for example ncurses.
Or specify the name of the library with --with-tlib.

Seems some terminal library is needed ...
11. The Ncurses (new curses) library

$ cd ..
$ wget http://ftp.gnu.org/pub/gnu/ncurses/ncurses-5.6.tar.gz
$ tar xvf ncurses-5.6.tar.gz
$ cd ncurses-5.6
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1
$ make
$ make install

10b. vim

$ cd ../vim71
$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-features=tiny
...
checking for tgetent in -lncurses... yes
ncurses library is not usable
checking for tgetent in -ltermlib... no
checking for tgetent in -ltermcap... no
checking for tgetent in -lcurses... no
no terminal library found
checking for tgetent()... configure: error: NOT FOUND!
You need to install a terminal library; for example ncurses.
Or specify the name of the library with --with-tlib.

seems we have to force it:

$ ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-features=tiny --with-tlib=ncurses
$ make
...
Starting make in the src directory.
If there are problems, cd to the src directory and run make there
cd src && make first
make[1]: Entering directory `/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/vim71/src'
CC="/opt/toolchains/uclibc-crosstools/bin/mipsel-uclibc-gcc -Iproto -DHAVE_CONFIG_H   -I/var/usb/usb_1        " srcdir=. sh ./osdef.sh
osdef0.c:10:44: termcap.h: No such file or directory
...
In file included from term.c:33:
/var/usb/usb_1/include/ncurses/termcap.h:64: error: conflicting types for 'tgoto'
...

hmm, I guess I really need to use the CPPFLAGS, shouldn't I modify the environment file? Perhaps, but now it's too late, it's finished:

$ make distclean
$ CPPFLAGS="-I/var/usb/usb_1/include/ncurses" ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1 --with-features=tiny --with-tlib=ncurses --disable-gui --disable-gtktest --disable-xim --without-x --disable-multibyte --disable-netbeans --disable-cscope --disable-gpm
$ make
$ make install
...
strip /var/usb/usb_1/bin/vim
strip: Unable to recognise the format of the input file `/var/usb/usb_1/bin/vim'
make[1]: *** [installvimbin] Error 1
make[1]: Leaving directory `/home/hajma/Dokumenty/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/vim71/src'
make: *** [install] Error 2

No, it's not finished... Let's just hack it:

$ diff -u src/Makefile.old src/Makefile
--- src/Makefile.old    2008-06-17 02:16:32.000000000 +0200
+++ src/Makefile        2008-06-17 02:16:47.000000000 +0200
@@ -999,7 +999,7 @@
INSTALL_DATA_R = cp -r
### Program to run on installed binary
-#STRIP = strip
+STRIP = echo
### Permissions for binaries  {{{1
BINMOD = 755
$ make install

I was quite surprised with the size of some files. Performing

/opt/toolchains/uclibc-crosstools/bin/mipsel-linux-sstrip
on bin/ and lib/ helped a bit, but 1MB of vim suggests there's still place for optimizations.



Final creation and writing of the squashfs image:

# /home/hajma/Dokumenty/asus/squashfs2.0r2/squashfs-tools/mksquashfs /var/usb/usb_1 /tmp/new.sqsh -be -noI -noD -noF
# dd if=/tmp/new.sqsh of=/dev/sda1

In the next part I'll try to correctly configure jabberd and make it start at router boot.



Customizing Asus AM200g - V. firmware modification

Customizing Asus AM200g - IV. jabberd configuration

Customizing Asus AM200g - II. Preparing the environment

Customizing Asus AM200g - I. PREPARATION

Customizing Asus AM200g - II. Preparing the environment

I know I promised you a jabber server, but in the end it was not so easy and the entry would be perhaps too long, so let's start with simple stuff :-)


Note I am not a programmer, have no C skills, so there can be a glitch here and there, but in the end all steps below led me to something that worked. If you find something really nasty, please let me know.

I took most of the inspiration at Daniel Fišer's page on Asus WL-600g hacking and just tweaked it to suit my setup.

1. On a fresh Mandriva 2008.1 installation some of the necessary tools are missing. Install them as root:

# urpmi make gcc autoconf automake

2. Since Asus still did not release the sources for AM200g, I used the GPL setup provided for AM604g, which uses the same chipset and seems to be almost identical (it even comes in the same box, except for missing USB port and second antenna). Follow these steps to install it:

$ wget http://dlsvr02.asus.com/pub/ASUS/Broadband/xDSL/GPL_AM604G_ForRussia.zip
$ unzip GPL_AM604G_ForRussia.zip
$ cd GPL_AM604G_ForRussia
$ tar xvf GENERIC_6348_WLAN_A34_3-06-02-01_GPL.tar.gz
$ su
# ./consumer_install

3. Not surprisingly, the AM604g source is just as broken as the AM600g one, so the same steps as described by Daniel need to be done:

# cd /opt/toolchains/uclibc-crosstools/bin
# for A in mips-linux-uclibc-* ; do
> ln -s ${A} mipsel-uclibc-${A#mips-linux-uclibc-}
> done
# cd /home/hajma/Documents/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01
# wget http://daniel.fiser.cz/IT/asus_WL-600g/cur_menu.patch
# patch -p0 <cur_menu.patch

4. Now you can try to build the firmware, just for fun:

# make PROFILE=GENERIC_6348_WLAN_A34

The firmware is created in the images directory. Since I have no replacement for the router, I'm not going to try it, although it might actually work.

5. To prepare the environment, create a file containing:

$ cat environment
TOOLCHAINS_HOME=/opt/toolchains/uclibc-crosstools
export LIBS="-L${TOOLCHAINS_HOME}/lib/gcc/mips-linux/3.4.2"
export INCLUDES="-I${TOOLCHAINS_HOME}/include/c++/3.4.2 \
-I${TOOLCHAINS_HOME}/mips-linux/sys-include \
-I/home/hajma/Documents/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/kernel/linux/include"
export CFLAGS="${INCLUDES} ${LIBS} -static"
export LDFLAGS="-static"
export CC="${TOOLCHAINS_HOME}/bin/mipsel-uclibc-gcc"
export LD_LIBRARY_PATH=${LD_LIBRARY_PATH}:${TOOLCHAINS_HOME}/lib
export PATH="${PATH}:${TOOLCHAINS_HOME}/bin"

and set the environment before doing anything else:

# . ./environment

(yes, this is "dot space dot ...")

6. Let's proceed to the helloworld example. Create it's own folder:

mkdir /home/hajma/Documents/GPL_AM604G_ForRussia/GENERIC_6348_WLAN_A34_3-06-02-01/userapps/opensource/helloworld

and follow the steps described in the automake manual:

Create the following files in an empty directory.

• src/main.c is the source file for the hello program. We
  store it in the src/ subdirectory, because later, when the package
  evolves, it will ease the addition of a man/ directory for man
  pages, a data/ directory for data files, etc.
  

  ~/amhello % cat src/main.c
  #include <config.h>
  #include <stdio.h>
  int
  main (void)
  {
  puts ("Hello World!");
  puts ("This is " PACKAGE_STRING ".");
  return 0;
  }
  

  
  
• README contains some very limited documentation for our little
  package.
  

  ~/amhello % cat README
  This is a demonstration package for GNU Automake.
  Type `info Automake' to read the Automake manual.
  

  
  
• Makefile.am and src/Makefile.am contain Automake
  instructions for these two directories.
  

  ~/amhello % cat src/Makefile.am
  bin_PROGRAMS = hello
  hello_SOURCES = main.c
  ~/amhello % cat Makefile.am
  SUBDIRS = src
  dist_doc_DATA = README
  

  
  
• Finally, configure.ac contains Autoconf instructions to
  create the configure script.
  

  ~/amhello % cat configure.ac
  AC_INIT([amhello], [1.0], [bug-automake@gnu.org])
  AM_INIT_AUTOMAKE([-Wall -Werror foreign])
  AC_PROG_CC
  AC_CONFIG_HEADERS([config.h])
  AC_CONFIG_FILES([
  Makefile
  src/Makefile
  ])
  AC_OUTPUT
  

  
  

Once you have these five files, it is time to run the Autotools to instantiate the build system. Do this using the autoreconf command as follows:

~/amhello % autoreconf --install
configure.ac: installing `./install-sh'
configure.ac: installing `./missing'
src/Makefile.am: installing `./depcomp'

At this point the build system is complete.
In addition to the three scripts mentioned in its output, you can see that autoreconf created four other files: configure, config.h.in, Makefile.in, and src/Makefile.in. The latter three files are templates that will be adapted to the system by configure under the names config.h, Makefile, and src/Makefile. Let's do this:

In addition to the manual's instructions, the details of host architecture must be provided when compiling for another platform. Also I set the prefix as the flash drive gets mounted as /var/usb/usb_1:

# ./configure --build=i686-pc-linux-gnu --host=mipsel-linux --target=mipsel-linux --prefix=/var/usb/usb_1/hello
# make
# make install

7. Optionally you can strip (remove all unnecessary bits) the binary:

# cd /var/usb/usb_1/hello/bin
# /opt/toolchains/uclibc-crosstools/bin/mipsel-uclibc-strip -o hellostripped hello

8. Copy the files to the flash:

# /home/hajma/Documents/squashfs2.0r2/squashfs-tools/mksquashfs /var/usb/usb_1 /tmp/dir.sqsh -be -noI -noD -noF
# dd if=/tmp/dir.sqsh of=/dev/sdb1

9. Plug the flash drive to the router and verify all runs fine:

$ telnet 192.168.1.1
Trying 192.168.1.1...
Connected to 192.168.1.1 (192.168.1.1).
Escape character is '^]'.
BCM96348 ADSL Router
Login: admin
Password:
> sh
BusyBox v1.00 (2005.04.12-18:11+0000) Built-in shell (msh)
Enter 'help' for a list of built-in commands.
# cd /var/usb/usb_1/hello/bin
# ./hello
Hello World!
This is amhello 1.0.
# ./hellostripped
Hello World!
This is amhello 1.0.

Final note: if you get "hello: applet not found" instead, it means you somehow did not compile the program for correct platform and BusyBox was not able to execute it.

Customizing Asus AM200g - V. firmware modification Customizing Asus AM200g - IV. jabberd configuration Customizing Asus AM200g - III. jabberd and vim Customizing Asus AM200g - I. PREPARATION

Customizing Asus AM200g - I. PREPARATION

As a fresh ADSL subscriber I became the owner of Asus AM200g ADSL modem/wireless router. The box is quite powerful - it is run by a 256 MHz processor, has 16 MB of RAM and is running Linux, so I'd like to use it for some other useful tasks. Since Asus has not yet released the source code it's not going to be that easy.



I want to touch the original firmware the least possible, so I'm going to add everything to an attached flash drive.
But the firmware only supports vfat and squashfs! Because of vfat filesystem limitations I decided to use a squashfs partition to store program files and a vfat one to store data that need to be modified, let's see how it works out. Note that all is done using Mandriva Linux 2008.1.

# urpmi squashfs-tools
$ mksquashfs /tmp/directory /tmp/dir.sqsh
Parallel mksquashfs: Using 1 processor
Creating little endian 3.1 filesystem on /tmp/dir.sqsh, block size 131072.
[==============================================================] 6136/6136 100%
Exportable Little endian filesystem, data block size 131072, compressed data, compressed metadata, compressed fragments, duplicates are removed
Filesystem size 4586.44 Kbytes (4.48 Mbytes)
8.61% of uncompressed filesystem size (53255.07 Kbytes)
Inode table size 53222 bytes (51.97 Kbytes)
23.13% of uncompressed inode table size (230058 bytes)
Directory table size 73178 bytes (71.46 Kbytes)
31.71% of uncompressed directory table size (230774 bytes)
Number of duplicate files found 3581
Number of inodes 7327
Number of files 6092
Number of fragments 224
Number of symbolic links  0
Number of device nodes 0
Number of fifo nodes 0
Number of socket nodes 0
Number of directories 1235
Number of uids 1
hajma (500)
Number of gids 0

Now I wrote the created image to the flash drive's first partition:

# dd if=/tmp/dir.sqsh of=/dev/sdb1

Rather surprisingly, the firmware only supports squashfs v2.0, so after plugging the flash to the router, it shows in the logs

kernel: SQUASHFS error: Major/Minor mismatch, filesystem is (3:1), I support (1 : x) or (2 : <= 0)

and in addition squashfs v3 does not support creating v2 images. Oh fsck. So I downloaded squashfs2.0-r2.tar.gz and

# urpmi make
# urpmi zlib1-devel
$ tar -xvf squashfs2.0-r2.tar.gz # note that tar now automatically gunzips the file too!
$ cd squashfs2.0-r2/squashfs-tools/
$ make
cc -I. -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -O2   -c -o mksquashfs.o mksquashfs.c
mksquashfs.c: In function ‘create_inode’:
mksquashfs.c:476: warning: integer constant is too large for ‘long’ type
mksquashfs.c:489: error: lvalue required as left operand of assignment
mksquashfs.c:541: warning: integer constant is too large for ‘long’ type
mksquashfs.c:546: warning: integer constant is too large for ‘long’ type
mksquashfs.c:549: error: lvalue required as left operand of assignment
mksquashfs.c:568: warning: integer constant is too large for ‘long’ type
mksquashfs.c: In function ‘write_file’:
mksquashfs.c:1030: warning: integer constant is too large for ‘long’ type
mksquashfs.c:1096: warning: integer constant is too large for ‘long’ type
mksquashfs.c: In function ‘dir_scan’:
mksquashfs.c:1275: warning: integer constant is too large for ‘long’ type
mksquashfs.c:1352: warning: integer constant is too large for ‘long’ type
make: *** [mksquashfs.o] Error 1
$

hmm, okay, the old crap does not want to be compiled with gcc 4 ...

# urpmi gcc3.3
$ CC=gcc-3.3.6 make
$ ./mksquashfs /tmp/directory /tmp/dir.sqsh
# dd if=/tmp/dir.sqsh of=/dev/sdb1

After plugging it into the router,

kernel: SQUASHFS: Mounting a different endian SQUASHFS filesystem on sda1

shows up in the logs. Okay, time to try

$ ./mksquashfs /tmp/directory /tmp/dir.sqsh -be
# dd if=/tmp/dir.sqsh of=/dev/sdb1

and I am getting one error message less! But I still see an ugly

kernel: SQUASHFS error: lzma_fs error while decompressing!
kernel: SQUASHFS error: Unable to read cache block [4bea97:7ae]
kernel: SQUASHFS error: Unable to read inode [4bea97:7ae]
kernel: SQUASHFS error: Root inode create failed" and it does not get mounted.

Googling reveals three links that do not help much. But since decompression is mentioned ... let's try to turn off compression ... :

$ ./mksquashfs /tmp/opends/src /tmp/dir.sqsh -be -noI -noD -noF
# dd if=/tmp/dir.sqsh of=/dev/sdb1

and voila, it gets mounted!

kernel: usb 1-1: new high speed USB device using address 7
kernel: scsi5 : SCSI emulation for USB Mass Storage devices
kernel: Vendor: USB 2.0 Model: Flash Disk Rev: 0.00
kernel: Type: Direct-Access ANSI SCSI revision: 02
kernel: SCSI device sda: 2047999 512-byte hdwr sectors (1049 MB)
kernel: sda: assuming Write Enabled
kernel: sda: assuming drive cache: write through
kernel: sda1 sda2
kernel: Attached scsi removable disk sda at scsi5, channel 0, id 0, lun 0
kernel: Attached scsi generic sg0 at scsi5, channel 0, id 0, lun 0, type 0
kernel: USB Mass Storage device found at 7
syslog: mkdir -p /var/usb/usb_1
syslog: mount /dev/sda1 /var/usb/usb_1

Next time I'll try to get a jabber server running, time to go sleep now. Customizing Asus AM200g - V. firmware modification Customizing Asus AM200g - IV. jabberd configuration Customizing Asus AM200g - III. jabberd and vim Customizing Asus AM200g - II. Preparing the environment

čištění displeje a dva obrázky z přírody

Můj Alcatel OT 535 je už dva měsíce po záruce, tak jsem si řekl, že se nic nestane, když ho rozeberu. Postupoval jsem podle návodu na mobilmanii a vyčistil jsem display od prachu.




Taknějak jsem doufal, že až ho dám zase dohromady, tak se spraví jedna věc, co mě štve. Asi před tejdnem jsem promazával smsky a asi mi během toho promazávání nějaká zase přišla a telefon se s tím nevyrovnal. Od té doby ukazuje ikonku nepřečtené zprávy furt. Vzhledem k tomu, že po zběžným pohledání na netu jsou všechny návody, manuály atd. za prachy, tak mi asi nezbude, než s ním někam zajít, aby mi ho zresetovali.

Telefon je to jinak relativně dobrej, i když mi přijde dost línej v lezení v menu. 1-2 vteřiny na zobrazení telefonního seznamu mi přijde dost. Hry stojej za houby a další jsou jen za $$. Javu to nemá, jenom nějakej brutálně proprietární systém, kterej není dostupnej ani developerům her. Ti maj k dispozici jenom soft.emulátor. Na druhou stranu přežil pád ze 3 metrů na dlažky. Ale stejně bych si příště radši pořídil nějakej jednoduchej černobílej - volání, sms, budík - nic jinýho s tím stejně nedělám, když nepočítám 2x použití GPRS. Hlavně aby to reagovalo hned a dlouho vydrželo na baterku.

Na ruku mi vlezla Klopuška mravenčí - Myrmecoris gracilis, s ulovenou mšicí.



Na lavičce na hřišti jsem si všiml dřevokazné houby, na netu jsem si dohledal, že se pravděpodobně jedná o kropilku - Dacrymyces minor.

Cisco Aironet 350 - konektor na externí anténu

Cisco Aironet 350 je velmi solidní pcmcia wi-fi karta se slušnou podporou v linuxu. Jediná její vada je v absenci externího konektoru, což se dá ale poměrně snadno napravit. Pozor, následující zápis obsahuje cca 500 kB obrázků.



Mou výchozí inspirací byl zápisek VK1EME, troufám si ale tvrdit, že můj přístup je elegantnější a v důsledku i praktičtější a hezčí.


Co je potřeba? Pájka (pokud možno nikoliv transformátorová, ačkoliv jednou jsem se zapomněl a přežilo to), cín, pájecí pasta, kleště, lámací nůž, epoxidové lepidlo, samičí konektor RSMA na 3mm kablík (hokej v R/SMA konektorech je pěkně vysvětlen zde) - v prodeji třeba tady, 3 mm wi-fi kablík - stačí pár centimetrů, chemoprén a kolíčky na prádlo.





Z karty je nejprve nutno sloupnout plastovou krytku, jde to snadno, stačí pod ni strčit nůž a zapáčit.





Ta věc vpravo vedle diodek je konektor U.FL. Jeho nevýhodou je, že je drahej, těžko k sehnání a dá se použít max 30x. Možná by bylo vhodnější zrušit ho a napojení externí antény udělat místo něj, ale VK1EME do toho nešel a já jsem nechtěl riskovat; koneckonců funguje to bez problémů.





Lámacím nožem teda vyřízneme plochu, kde budeme operovat. Příště už bych nechal vlevo trochu víc místa, špatně se mi pak lepil kryt zpátky.








Ten zlatej obdélník je samotná vnitřní anténka. Aby bylo možné připojit externí anténu, je nutný ji odříznout od elektroniky karty, mělo by stačit oddělení cca 1mm, ale čím víc, tím líp. Vývody z karty je potřeba ošmirglovat, aby se potom dobře pájely - ale opatrně, aby se odšmirglovanej kov nedostal tam, kam nemá.





Na kablík se konektor přidělá klasicky, zapájí se středový pin,protáhne, přes opletení se přetáhne trubička a zamáčkne kleštěma, aby to bylo placatý.








Druhý konec kablíku hned za konektorem zkrátíme, odizolujeme střed a stínění rozdělíme na dva "copánky", pokud možno v rovině se zmáčknutím konektoru.





Konektor pak (opatrně) připájíme tak, aby byla jeho nejtlustší část v okraji karty. Středový vodič do prostřední cestičky, stínění na kraje.
(Nevím, jestli je to nutný, ale pro jistotu jsem ještě tu původní anténku přelepil izolačkou.)








Nakonec se konektor zaleje ze všech stran epoxidovým lepidlem a nechá zaschnout.





Originální umělohmotný kryt karty je potřeba trochu upravit - vyříznout v místě konektoru, který je přeci jenom trochu tlustší. Pro nalepení nazpátek jsem si pomohl trochou chemoprénu, secvaknul kolíčky na prádlo a nechal zaschnout.


Et voilà, zde je výsledek:

Cisco Aironet 350 - connector for external antenna

Cisco Aironet 350 is quite good pcmcia wi-fi card well supported in linux. It's only flaw is the absence of external connector, which can be easily resolved.


My initial inspiration was an article by VK1EME,
but I dare to say that my approach is more elegant and in the end more practical and prettier.



What is needed? Soldering iron (not the one using transformer, if possible, although I have forgot about it once and the card survived), solder, flux, pliers, knife, epoxy glue, female RSMA connector for 3mm cable (the chaos in R/SMA connectors is explained in depth here) - sold e.g. here, 3 mm wi-fi cable - just a few centimetres, glue and clothes pegs.





First you need to strip the plastic cover, it goes easily, just stick a knife below it and lever.




The thing next to the diodes is a U.FL connector. It's disadvantage is cost, is difficult to get and one can use it only 30 times. Perhaps it would be be better to remove it and plug the external connector instead of it, but VK1EME didn't and I didn't want to risk; it works ok, so why bother?





Use a knife (the kind that is used for cutting cardboard) to make space for further operation. Next time I would leave more plastic at the left to ease glueing the cover back.








The golden rectangle is the inner antenna. It is necessary to cut off its connection with the inner card electronics, 1mm space should be enough, but the more the better. The leads need to be sand-papered for better soldering, but be careful - the metal dust shouldn't get inside.





Tag the connector on to the cable in the usual way - solder the middle pin, put it through, put the tube around the shielding and press with pliers to make it (a bit) flat.








Cut the other end of the cable right near the connector, strip the insulation and the middle part and divide the shielding into two "plaits", if possible on a level with the connector.





Then (carefully) solder the connector so that its widest part was at in the edge of the card. The middle wire to the middle lead, shielding to the border ones. (Not sure if it's necessary, but to be sure I covered the inner antenna with a tape.)








Then seal the connector on all sides with the epoxy glue and let it dry.





The original plastic cover needs adjustation - cut out where the connector is. Then use some glue, fix it with the pegs and let dry.


Et voilà, here's the result:

Debugování komunikace s klávesnicí

Momentálně si hraju s vylepšováním svojí klávesnice, akorát mi to moc nejde. Už jsem strávil tejdny různým zkoušením (no, nemám na hraní celý dny, proto tolik), až jsem uznal, že je nutný vědět, co se mezi klávesnicí a PC vlastně děje, protože podle teorie to prostě nefungovalo.


Takže jsem vyštrachal obvod MAX232, 5 kondenzátorů a napojil klávesnici na sériový port.







Šmírovací program jsem obšlehl od sitha. V C totiž vůbec neumím a připadá mi jako totálně nepohodlný a nepřehledný jazyk, takže - vím, že je to napsaný prasácky, ale je to jednoúčelová věc a stačí mi:

#include <stdio.h>
#include <fcntl.h>
#include <termios.h>
#include <sys/ioctl.h>
#include <time.h>
#define SERIAK "/dev/ttyS0"
int openserial ();
void closeserial ();
float mereni ();
int fd;
int
openserial ()   //otevrit seriak
{
if ((fd = open (SERIAK, O_RDWR)) == -1)
{
return 0;
}
return 1;
}
void
closeserial ()   //zavrit seriak
{
close (fd);
}
float
mereni ()   //hlavni fce, mereni napeti
{
int status, hodnota1 = 2, hodnota2 = 3;
while (1!=2)
{
ioctl (fd, TIOCMGET, &status);
hodnota1 = (status & TIOCM_CTS)/TIOCM_CTS; //clock
hodnota2 = (status & TIOCM_RI)/TIOCM_RI; //data
printf ("%i,", hodnota1);
printf ("%i\n", hodnota2);
}
// return (hodnota1);
}
int
main ()
{
if (!openserial ())
{
printf ("Nelze inicializovat port \"s\".", SERIAK);
return 0;
}
mereni();
closeserial ();
return 0;
}

Prostě to vypisuje dvě hodnoty oddělený čárkou, první je stav CLOCK a druhý je stav DATA. Protože PS/2 jede jen na 9600 baudů, stíhá se to krásně:

...
1,0
1,0
1,0
0,0
0,0
0,0
0,0
0,0
0,0
0,0
0,0
0,0
0,0
0,0
0,1
0,1
0,1
...

Update:

A když se to nacpe do OpenOffice, tak se to dá krásně zvizualizovat:






English summary: Spying on the PS/2 keyboard protocol using serial port.

Zas nic

Už jsem sem dlouho nic nenapsal, nějak není čas, a to tak, že vůbec na nic, teda možná je, jenom se snažím dělat deset věcí naráz a navíc se vždycky objeví nějaký zádrhel.


Jako třeba teďka, hraju si s projektem, který zahrnuje určité úpravy standardní AT klávesnice (detaily zatím neřeknu, ale je to super věc), takže se mi podařilo najít na netu dva projekty, který dohromady dají to, co chci, strávil jsem tejden čtením katalogovýho listu mikrokontroleru 16f84a a chápáním, jak se dělá v asssembleru, postavil jsem si programátor a když jsem se po pár pokusech dostal k smysluplnýmu kódu, tak ten programátor prostě přestal fungovat.
Podezřívám z toho nějakou změnu, co proběhla v Cookeru, protože se objevily i problémy s digitálem, konečně na tom programátoru není co spálit a sériovej port je snad docela jištěnej. Takže teďka zase budu zjišťovat co by mělo jak kde bejt, aby to fungovalo nebo na kus disku nainstalím LE2005 nebo jiný distro.

update - jo a taky mi už asi tejden sem tam ustřelí myš, že prej lost synchronization, to bude asi někde v kernelu...

T230 podruhé

Tak se mi to nepodařilo spravit, i když jsem zlikvidoval konektor. Doufal jsem, že chyba (zkrat) je tam, ale to jsem se asi pletl :-( Mobil si prostě pořád myslí, že je k němu připojený handsfree. Nakonec jsem k němu jako nouzovku přidělal sluchátko (s vlastním reproduktorkem si to nerozumí, asi je tam jiná impedance či co), aspoň něco než pořídíme něco vodotěsnýho :-)

T230

Malej si rád hraje s T230 mojí ženy. Bohužel jeho hry jsou pořád tak trochu slintací, což nědělá dobře jemné asijské elektronice, a tak se telefon jednoho dne odmlčel, resp. přestal fungovat zvuk.


Bylo mi jasné, že s takovým nánosem měděnky nemáme stejně na reklamaci šanci :-), takže jsem ho otevřel, odšrouboval 4 viditelné šroubky a přemýšlel co dál. Nakonec jsem zjistil, že zbývající půlkryt se otevře tak, že se uprostřed lehce zapáčí (cca 1 mm nahoru a pak vytlačit dopředu).


Tak se objevily další dva šroubky, ale pořád to ještě nebylo všechno. Poslední spoj je totiž lepený, je potřeba odlepit bílou folii, která přidržuje základní desku.

Vzal jsem teda líh (isopropylalkohol by byl lepší, vím) a vatičky a likvidoval měděnku. Měděnka je teď sice pryč, ale s mobilem je to pořád stejný. Mám podezření na konektor, kterým se připojuje handsfree, takže ho za chvilku asi zlikviduju...

Silent Mouse - The Second Encounter

Having used my silenced rodent for nearly 1/2 year at home, I finally decided to get rid of the clicking noise at work too and here's the how-to. Some people wanted me to translate the original article, but I guess it's better to start afresh with a revised and simplified approach.


I'm not going repeat reasons for silencing mice, if you want to do it, you probably know why. If you don't know why, then just assume there are people who hate the clicking sound.

Equipment needed:
a Genius Netscroll+ mouse - my original object of experimentation

or other type, such as my Dell scroll mouse (seems to be a Logitech)

(these two have been proven to work)
cross-point screwdriver
soldering iron
tin-lead solder
rosin or flux
sharp knife
2 cm of thin insulated wire (2 pcs) (only if necessary)

1. Disassemble the mouse. Now you can see the three switches that make more than 90 % of the noise (the 10 % are made up by resonance in the plastic body - I tried to fill it with some silencing material, but got no audible difference).

Take a look at the switches, they must look more or less like this one:


My method is only possible with this type of switch, if you see other type in your mouse you'll probably have to invent something else. This is the case of middle button of Dell mouse (green circle above).


2. Take notes on the position of white switch buttons. This step is probably unnecessary, as this seems to be always marked on the pcb's.



3. Carefully solder out the two outer switches. In my first attempt I silenced also the middle switch, which turned to be a bad solution. The switch became too sensitive and could be triggered by mere scrolling. I use the middle button rarely, so I left it noisy in the end; but if you do use the middle button often and therefore wish to hush it too, try it, you have been warned.



4. Open the switch by carefully lifting the legs (perhaps there's better word for it) on its sides.

It's wiser to put the switch upside down beforehand, so that the tiny white plastic button won't fall out. It's not a serious problem if you manage to break the legs - just glue the parts together afterwards - or forget it completely and take the risk of mouse malfunction in extreme conditions (e.g. earthquake).



5. Turn the switch cover by 180 degrees and join it with base again.

before: after:

What is the purpose? This way the point where the button touches the spring is shifted and somehow the spring now moves gradually rather than at once.
The switch lids of the Dell mouse were a bit different, there wasn't enough room to fit the switch after rotation, so I had to make more space with hot solder gun.



6. Solder the switches back to the PCB. They have to be in the same position as in point 2 - if the button was in front, it has to be in front again...



7. In case where only two pins are connected to the circuit (Genius), the pins of the switches are now reversed. To fix it, just connect the outer pins with a piece of wire. (note that this is an old photo, the middle button is no longer silenced and there's no wire there now)

With other type of mouse double check the schematics on how the pins are used.

The Dell mouse has both outer pins grounded, so there's nothing left to do.




8. Reassemble the mouse and enjoy.


You can check recorded sound of clicking before and after. FYI, the sounds are of the Genius mouse, but trust me, the other one is dead silent too, I was just too lazy to record it.

Side notes:
The disadvantage is, that the actual clicking is perhaps not as distiguishable or maybe accurate (no audio feedback in fact), but it's just a matter of getting used to. Other ways I have tried consisted of shortening the distance the spring has to travel, but the noise reduction gained was not satisfactory for me..
Also the mouse buttons are now more sensitive, so you'll get a few false clicks in the first couple of days, before you learn how to handle it.

Co s myškou ...

Pořád na mě z poličky smutně kouká stará rozkuchaná sériová myška, který jsem zničil mikrospínače, když jsem zkoušel, jak se dá ztišit zvuk klikání.

Tak mě napadlo, že jestli budu mít někdy čas, náladu, nástroje, materiál a tak, zkusím z ní udělat meteorologickou stanici ;-)

Anemometr udělám tak, že vrtulku protáknu tím kolečkem co přerušuje prostor mezi fotodiodou a infraledkou. Teploměr by mohl bejt jednoduchej obvod z termistoru a 555, kterej bude simulovat klikání tlačítka. No a jako soft postačí ovladač myši ;-)))