Kudelski N3 bits and pieces, plus thoughts on key / rsa extraction from flash.
N3 Notes mostly from forum posts by TheCoder

These are notes for future reference more than anything else, so please no excited emails about how its wrong, or right, or can I hack your box.

—
There are three different pairing methods used N3 boxes presently. These are DT06, DT08 and Secondary key.

The DT06 method transfers a compressed form of an rsa pq keyset from which the CAM public/private rsa keyset and its associated modulus can be derived.

The DT08 method transfers the cam modulus along with the IRD number of the married box. The public rsa key is not transferred but it is implied that the box already knows this value.

The Secondary key method does not involve a transfer. It imples that the box already knows the cards matching CAM modulus and rsa public key value.

Various boxes, depending on make/model, may use any of the above pre-pair key transfer methods but it could be useful to know which box uses which method.

http://www.digital-kaos.co.uk/forums/f10/people-n3-cards-82421/

Instructions:
1 Stick your N2/N3 card in your card reader
2 Run NagraEdit – DO NOT ATTEMPT TO READ YOUR CARD !!!
3 Select the Comm Tab. This should give you an upper and lower text pane
4 Cut/Paste the scriptt below into the top pane
5 Press the “Send D2C” button/icon
6 Results should appear in bottom pane
7 Interpret your results based on info below.
Script – Read DT06/DT08
Code:

rs
tx 21 C1 01 FE 1F
rx
tx 21 00 08 A0 CA 00 00 02 12 00 06 55
dl 02 00
rx
dl 02 00
tx 21 00 09 A0 CA 00 00 03 22 01 00 1C 7E
dl 02 00
rx
dl 02 00
mg *
mg *** DT06 info ***
tx 21 00 09 A0 CA 00 00 03 22 01 06 13 **
dl 02 00
rx
mg DT06 response1
dl 02 00
tx 21 40 09 A0 CA 00 00 03 22 01 86 13 **
dl 02 00
rx
dl 02 00
mg DT06 response2
mg *** End DT06 info ***
mg *
mg *** DT08 info ***
tx 21 40 09 A0 CA 00 00 03 22 01 08 13 **
dl 02 00
rx
mg DT08 response1
dl 02 00
tx 21 00 09 A0 CA 00 00 03 22 01 C8 55 **
dl 02 00
rx
dl 02 00
mg DT08 response2
tx 21 40 09 A0 CA 00 00 03 22 01 88 55 **
dl 02 00
rx
mg DT08 response3
dl 02 00
mg *** End DT08 info ***
*******
Just to clarify :
The important bits your looking at are the DT06/DT08 responses (the bits that start with Rx: )
ie RX: 12 00 15 A2 11 08 E0 00 00 00 5E 01 20 00 00 00
00 00 00 00 00 00 90 00 B3
RX: 12 40 15 A2 11 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 90 00 64
RX: 12 00 15 A2 11 00 00 00 00 00 00 00 00 00 00 00
00 00 00 00 00 00 90 00 24
and
RX: 12 40 57 A2 53 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 00
00 00 00 00 00 00 00 00 90 00 64

If the responses vary significantly from the above, with the 00’s replaced with some varying data, then its likely your card had the specified tier and is probably using the corresponding pairing method.

If the DT06 response contains lots of 00’s then its NOT DT06 pairing
If the DT08 response contains lots of 00’s then its NOT DT08 pairing
If its not DT06 or DT08 then its probably secondary key.

For non DT08 cards (mostly the newer boxes) each box has a unique cam_n already built into its firmware – this can only be extracted from the actual box itself.

About the Algorithm,

N2 encrytion was based on the following algorithm ->
decrypted message = ( (IDEA( ( (EncryptedMsg ^ 3)Mod N1) ,IdeaKey) ) ^3) Mod N1
Thats 2 distinct algorithms. An RSA algorithm (performed twice) and an IDEA algorithm.
The key for the RSA algorithm is something like –
RSA EMM-G=15A811B2065DF39CD48C9C958E7B406345295B09D0E9A18A 9B92C5FD7761CAAFAB830880F1F06B4E4477F157EA10D0AFC3 FDDB1ED2E7E83E89F03FF81237047DB76F79D6A2CFD75A7255 D72E52E7F47B96C2DFBDEBFC80CE927F6AD351FDF0BF8DA13F F62295BFBAF29035A230136D0B4AA99D38DD8B0465F2C709FC 8818173C
Which, as you can see, is a 128 byte (1024 bit) number. This is the main encryption key for EMM decryption.
The IDEA algorithm, which acts as an inner layer to the RSA has a standard 16 byte (128 bit) IDEA key.
There is no algorithm in N2 (or N3) that only uses an 8 byte (64 bit) key although some providers have opted to use 3DES rather than IDEA as an inner layer. This uses three separate 8 bye (64 bit – only 56 bits used) keys to form an amalgam 168 bit DES algorithm.

If Using DT08 (0a) on the card :
The Dt08 (0a datatype on card) is created by the provider and sent to the card at sub time.
The dt08 contains the Cam N public rsa key along with ird/boxkey.
The dt08 is IDEA encrypted with the Idea Key made from ird/boxkey/inverted ird.
The dt08 is RSA encrypted using Ird N (public rsa key) and Ird D (private key and uknown by anyone but provider).
Ird N = N1 xored N2
Ird N1= A4E9B585932F90282FD70C908176E8605E6B2CE629335A0FC1 5B31DAB0BFC6FEEB88CFC69649994CD3FE039C9965C620C4D5 828E9153998EE4AE0E8C25644DF3 xor
Ird N1= 237280AAB36BE4B21FC71FBF08218E532A545E744D7B007FF8 69BA426831C4AC653F3825ADE9358FCD1F0239EC447CBC2765 CC0AEBE437AF2270FC461C2FA042
Ird N = 879B352F2044749A3010132F89576633743F729264485A7039 328B98D88E02528EB7F7E33BA0ACC31EE101A57521BA9CE3B0 4E847AB7AE21C6DEF2CA394BEDB1
The Ird N1,N2,Ideakey exist in the tsop.
Ird E = 3
Ird D = UKNOWN, this is the reason you can’t create your own dt08 without changing the N1/N2 on the tsop, you must know Ird D.

DT08 (0A) = IdeaEncrypt(CamN/Ird#/Boxkey/Idea Signature,Ird_Ideakey) ^ D mod Ird N.

Ird requests DT08.
Card sends back the dt08 (0a)
Ird decrypts the dt08.
Decrypted dt08 = IdeaDecrypt(DT08,Ird_IdeaKey) ^ 3 mod Ird N.
It checks the ird # and boxkeys in the Decrypted 08 if they match what is on ird,
it stores the Cam N in the decrypted 08 in ird memmory.

If Using Secondary Key (SK) on the Ird.
Ird checks for SK exists on the ird, if it does, the dt08 will never be requested/ignored from the card.
Ird validates the SK with idea signature in the SK (using IIIIIIII01924314051647990A9C4E1 where I = irdnumber).
Ird takes the Cam N in the SK and puts it in ird memmory
Note : Cam N is not even encrypted in the sk, very weak method compared to dt08.

Later, establish session key (0C datatype on the card):
Ird requests 2a data from card.
Random 2a is sent from card to ird.
Ird performs some Idea signing (leave it to you to look up 2a/2b routines)
Ird comes up with session key from the 2a message sent from Cam.
Ird encrypts the session key with rsa.
Encrypted 2B = (2B data with 16 byte session idea key) ^ 3 mod Cam N.
Sends encrypted data back to card in 2b message.
Cam decrypts 2B with Cam N, Cam D. Decrypted 2B = (Encrypted 2B) ^ Cam D mod Cam N.
If valid, store session key in ram and on card for later use.

This all happens as ird boots.
When you select a channel.
Ird sends Cmd 07 ECM message with control words encrypted.
Cam decrypts the control words rencrypts them with Idea encryption using the session key established above.
The ird then requests the control words.
The Cam sends them back in the 1C response.
The ird decrypts the control words with with Idea encryption using the session key established above.
Sends the control words to the mpeg decoder.
8 seconds of video.
Repeat 07/1C process over and over.

——–

and pay special attention on CMD$2A ??

it should reply the CAMID serial number + 64 bytes random key generated and then shortly after encrypted with the CAM(AKA Conditional Access Module or SmartCard) RSA primary 96 bytes from the card… the so famous RSA modulus keys 96bytes at eeprom $8xxx on the ROM110 days.. From the card (this cmd is also the first step of the SessionKey negotiation)

shortly after the receiver receives this card reply.. and will decrypt it using the Secondary Key which is also 96bytes, this key is build up by using the following information stored in the receivers flash..

IR IR IR IR ZZ ZZ ZZ ZZ ZZ ZZ ZZ ZZ 00 03 F1 F1
F1 F1 F1 F1 F1 F1 SK SK SK SK SK SK SK SK SK SK
SK SK SK SK SK SK SK SK SK SK SK SK SK SK SK SK
SK SK SK SK SK SK SK SK SK SK SK SK SK SK SK SK
SK SK SK SK SK SK SK SK SK SK SK SK SK SK SK SK
SK SK SK SK SK SK F2 F2 F2 F2 F2 F2 F2 F2 CR CR

total 96 bytes..

IR = Receiver serial number
XX = Unimportant
EE = RSA public exponent for STB
F1 = SK signature 1 used to calculate the box key
F2 = SK signature 2 used to calculate the box key
SK = RSA public modulus N
CR = CRC Checksum
BB = BoxKey result from xoring F1 with F2 keys stored in the flash firmware from the receiver

once we decrypt the cmd2A we extract from inside the original 64byte random key generated in the card, then we will apply the IDEA encryption algo on the first 32bytes from the 64 byte random key to hash out the session key.

So this first 32 bytes are extracted from the 64byte random key and will be encryted using the IDEA SIGNATURE key… this key will be generated by the following information

IdeaKey generation
BB BB BB BB BB BB BB BB + CC CC CC CC + CA MI DC AM

BBBBBBBBBBBBBBBB = Boxkey result from F1 xor F2
CCCCCCCC = IRD number from receiver stored in Flash firmware
CA MI DC AM = CAM ID or Smart Card serial number converted in HEX, which can also be extracted by simply sending INS CMD$12 to the card..

so the IDEA signature key for encrypting the first 32bytes extracted from the 64 random seek key is

BBBBBBBBBBBBBBBBCCCCCAMIDCAM

once applied the IDEA encryption we will have the result 16 byte sessionkey.. which will be stored in the receiver flash for a few hours… to be more precise around 5 hours..

Now going back to the calculation done before, the receiver decrypted the cmd$2Aencrypted by the card with the RSA primary 96 stored in the card.
once decrypted it simply just extracted the first 32bytes of the 64 byte seed key generated by the card. this 32 bytes were used for calculation of the 16byte sessionkey..

but, before the 32byte keys was taken for the idea signature encryption.. the receiver.. re-encrypted the 64byte random key using the SECONDARY KEY RSA 96 stored in the receiver flash, which i just stated above how to get this key…
and will send it back to the card on cmd$2B

the card will receive the cmd$2b and will decrypt it using the PRIMARY RSA modulus key 96.
and will extract just the first 32bytes.. and by using the BOXKEY+IRD+CAMID stored in the card, the card will also calculate the same 16 byte session key.

in orde to make the card pairing , u need to know the RSA_N+BOXKEY+IRD NUMBER+CARD SERIAL number or CAMID…

then with them all together we can start comunication between the card.. and on the CMD$2A and $2B we will have the SessionKey negotiation which i just explained previously.

if for example on one side or the other we have different BKand RSA… we will a session key failure.. and without this Sessionkey we will not be able to decrypt the CM$1C or $9C related stuff

this means that if negotiation of session key succeds, then the receiver will send the ECM$07 to the card, which will then be decrypted by simply just using the ECM RSA modulus key and the ECM IdeaKey to decrypt the Control Words / CWs.

once they are decrypted the card will send them to the RECEIVER.. this is normally done via CMD$1C obviously this CMD is encrypted with the Sessionkey 16 BYTES described above.. once it arrives at the Receiver end, you will use the same Sessionkey 16byte to decrypt that CMD$1C and extract the REAL DCWs Decrypted Control Words to open up the Video and Audio stream related for that XX amount of seconds…

Now this Session key is refreshed every 5 hours.. this means that every 5 hours a new session key is produced.. so every 5 hours the card generates another 64byte random seed key using other RSA algo inside the card.., and will then send this 64 byte seed key to the receiver again.

Shortly followed by all the procedure described above to extract the new session key again.

—————-

Given that ==================================================
SK 96 BYTES
==================================================
11111111<----------------------------FIRST 4 BYTES---- IRD XXXXXXXXXXXXXXXXXXXX<--NEXT 10 BYTES---- unknown 1111111111111111<--------8 BYTES---- Y1---WRITE DOWN ================================================== 11111111111111111111111111111111-_ 11111111111111111111111111111111-_-_"N" 64 Bytes 11111111111111111111111111111111-_- 11111111111111111111111111111111- ================================================== 1111111111111111<-------8 BYTES--- Y2-----WRITE DOWN XXXX<-------------------2 BYTES--- CHECKSUM ================================================== Y2 Xor Y1 = BOXKEY ================================================== 006e convert from hex gives you 110 bytes block cipher ird 4 bytes bk 8 bytes sk 96 bytes leaving us with 02bytes used to encrypt, decrypt above. eg: 00 00 00 6e 34 0d 20 1c 03 03 70 80 5a 8e dd 24 ac cc a4 a6 e2 da 86 91 29 18 0b a6 23 6d fa c4 05 7f 1b 20 97 eb 0c 19 b3 39 2f 1e cb 9b 67 4d ed 10 f5 65 ec 0d c7 35 ac f0 b8 89 b0 51 59 22 69 85 d5 f1 93 48 7a 84 6e 1f b4 24 83 79 db 02 4d b0 9c 5e 8b df 89 57 9c 5a 7f 9a cc 87 51 3b 15 6b 15 cc c4 2f 66 e7 e6 75 4f 24 f2 07 85 0d db b0 3d e2 64 dd e9 54 ad 77 60 e7 8f a6 cd a6 46 c3 b8 fa e4 e7 51 2d 6a 2f 95 68 56 b5 78 34 17 6b b8 48 38 87 c4 95 e5 b0 41 2c 95 e1 24 aa 4b 2a 6f 8c 90 53 29 a9 6b 3d 0a b5 92 1c 95 ec 72 b9 54 a9 99 f5 f3 dd f4 0f 60 c3 25 5b 5b 81 22 e8 79 c5 be 8f 3c 89 2b 8a ad ba 27 b0 c2 f7 b1 4f 08 d5 37 2a 97 c5 f0 07 9d 99 be c7 8a a9 cf 5a c5 45 ce 1e 25 43 81 95 7a 22 33 ed 93 74 Gives: 00 00 00 63 (length) 0d 20 1c 03 03 70 80 5a 8e dd (unknown) 24 ac cc a4 a6 e2 da 86 (y1) 91 29 18 0b a6 23 6d fa c4 05 7f 1b 20 97 eb 0c 19 b3 39 2f 1e cb 9b 67 4d ed 10 f5 65 ec 0d c7 35 ac f0 b8 89 b0 51 59 22 69 85 d5 f1 93 48 7a 84 6e 1f b4 24 83 79 db 02 4d b0 9c 5e 8b df 89 (64 bytes) 57 9c 5a 7f 9a cc 87 51 (y2) Y2 XOR Y1 = 0x579c5a7f9acc8751 xor 0x24accca4a6e2da86 = 42 12 8C F2 39 39 55 FA So, a flash dump would be helpful.. Possible Scenario's - BGA, TSOP, TLGA etc.. desoldering for Flash. Put in a socket. eg from http://shop37051047.taobao.com/ or http://shop34694309.taobao.com/?search=y

Pop flash back onto something else, read, dump. eg
http://item.taobao.com/item.htm?id=7422440993

Get box key, rsa key (if req. based on a check of DT type from the actual subbed card)
Pop flash back in socket.

Been there, done that for data recovery on faulty flash drives, plus most of the places I know down at QJlu have SMD / BGA desoldering capability or better.

Bunnies blog is fairly good at explaining the basics (albeit for xbox) – http://www.xenatera.com/bunnie/proj/anatak/xboxmod.html

Most boxes use ARM based SoC’s for things. Also possible to just throw up a dev board, and interface to that.

Although most boxes also have some form of OS running, so just as feasible to dump flash that way also assuming serial or jtag access and a bootloader is available.

Amusing that people like http://www.flashbackdata.com/blog/?p=195 claim this is hard – there are plenty of tools for this already out there eg softcenter, pc3000, plus all the local chinese stuff. Semi ok forum here talking about flash recovery, although not as technical as I’d like.

http://forum.hddguru.com/hard-disk-drives-english-forum-f12.html

Once key(s) are had, then I can use my own decoder rather than the crappy one the broadcaster uses.

Yay.

Had a client come into the office today with a locked iPhone.

Normally this isn’t really a big deal (assuming that there is a hack for it), but in this case, it was a little more complex, as he didn’t have working wifi.

PwnageTool has a great feature where you can add Cydia Packages to a custom firmware, so that you can prepackage the firmware already to go.

So, I opened up PwnageTool, added the http://repo666.ultrasn0w.com/ site to the Cydia sources section in advanced, and tried to load in Ultrasn0w.

Life isn’t easy, and it didn’t work.

But why didn’t it work?

I took a look at a working site, and checked out the differences between their package section and Ultrasn0w’s.

Ultrasn0w is hosted on repo666.ultrasn0w.com
While their website doesn’t really tell you much useful information, a bit of googling lead to some info.

The .deb file (debian package file) on their site is at http://repo666.ultrasn0w.com/ultrasn0w.deb

Cydia usually needs stuff in a particular format, so I next checked out how one makes a repository.
This is documented at Saurik’s site here – http://www.saurik.com/id/7

Basically, you throw files into a folder and make a Packages file.

The example given on Saurik’s site is this:

/web/apt/xmpl]# dpkg-scanpackages -m . /dev/null >Packages
** Packages in archive but missing from override file: **
com.saurik.myprogram

Wrote 1 entries to output Packages file.
[root@desktop:/web/apt/xmpl]# bzip2 Packages
[root@desktop:/web/apt/xmpl]# ls -la *
-rw-r--r-- 1 root root 906 2008-07-01 07:48 MyProgram.deb
-rw-r--r-- 1 root root 380 2008-07-01 08:00 Packages.bz2
[root@desktop:/web/apt/xmpl]#

So, it appears we need a Packages.bz2 file.

Being adventurous, I decided to setup my own repo, and stuck the .deb file for Ultrasn0w in there.
Followed the instructions and created the Packages.bz2 file.

Tried again in PwnageTool, and… No go.

Hmm.

Does http://repo666.ultrasn0w.com have a Packages.bz2 file?
Why yes it does.

Take another look at the working one – ahah says my brain.

They point the folder to the _uncompressed_ Packages file.
I guess PwnageTool doesn’t support compressed Package list files.

So, I try that out using an uncompressed file.
Created the Packages file with

dpkg-scanpackages -m . /dev/null >Packages

and try again.

Better – I’m getting a result now with my repo when I click refresh.
However, I can’t seem to be able to download any files…

So, lets take a look at whats happening in my apache logs.


58.37.213.199 - - [07/Mar/2011:20:50:52 +0800] "GET /dists/Packages HTTP/1.1" 200 1643 "-" "PwnageTool/4.2 CFNetwork/454.11.5 Darwin/10.6.0 (i386) (iMac9%2C1)"
58.37.213.199 - - [07/Mar/2011:20:51:06 +0800] "GET /./mobilesubstrate_0.9.3228-1_iphoneos-arm.deb HTTP/1.1" 404 1184 "-" "PwnageTool/4.2 CFNetwork/454.11.5 Darwin/10.6.0 (i386) (iMac9%2C1)"
58.37.213.199 - - [07/Mar/2011:20:51:42 +0800] "GET /./ultrasn0w.deb HTTP/1.1" 404 1164 "-" "PwnageTool/4.2 CFNetwork/454.11.5 Darwin/10.6.0 (i386) (iMac9%2C1)"


Aha! While its successfully found the repo now, its looking for the files in the wrong folder – my repo is in /dists, and its looking in the root folder.

Seems the Saurik instructions are a bit mangled, or the Package generator is a bit silly.
Quick look at the helpfile shows it needs the folder via -m

So I went up a level, and regenerated my file.


cd ..
dpkg-scanpackages -m dists > dists/Packages


Yes, it works!

Now PwnageTool can download my file finally. Yay!

I just need to select it in PwnageTool / Packages as below, and build my ipsw to test.

Now I can finally make my own Ultrasn0w firmware woohoo!
Not as hard as it seems, but not as easy either!

I’ll leave my UltraSn0w repo at http://www.sheed.com/dists/ for now, but will probably move it elsewhere at some point, and update this post. So, if you need it, get it while you can.

Lawrence.

This was posted by myself over at BimmerForum’s, but as China’s Firewall is being particularly unpleasant these days about what sites I can and cannot visit, I thought I’d repost it here, so its more accessible for me.

Note – I no longer own a 7 series, so any requests for modded rom’s should probably be accompanied with nice gifts to persuade me to do the work involved 🙂

As I saw people asking about Tuned ROM’s in another post I thought I’d do some research.

Its quite niche – hence the ridiculous pricing people charge for roms – time is money though!

Here’s my research so far. I’m basing it on the E32 735iL 1989/1990, as thats what I own (and can test for).

Googling for E32 735iL gives:

ELECTRONIC CONTROL UNIT, DME MOTRONIC

BOSCH 0261200179
BMW 1730697
*26SA0840*

Vehicle Applicability:
1986-1993
BMW 3′ E30 320i Touring 2.0 95kW/129PS (M20/ 206EE/KA)
BMW 5′ E34 535i 3.4 155kW/211PS (M30/346KB)

BMW 6′ E24 635CSi 3.4 155kW/211PS (M30 B35)
BMW 7′ E32 735i 3.4 155kW/211PS (M30/ 346KB/EC)
BMW 7′ E32 735iL 3.4 155kW/211PS (M30/ 346KB/EC)

ELECTRONIC CONTROL UNIT, DME MOTRONIC

BOSCH 0261200179
BMW 1722611
*26RT2794*

Vehicle Applicability:
1986-1993
BMW 3′ E30 320i Touring 2.0 95kW/129PS (M20/ 206EE/KA)
BMW 5′ E34 535i 3.4 155kW/211PS (M30/346KB)
BMW 6′ E24 635CSi 3.4 155kW/211PS (M30 B35)
BMW 7′ E32 735i 3.4 155kW/211PS (M30/ 346KB/EC)
BMW 7′ E32 735iL 3.4 155kW/211PS (M30/ 346KB/EC)

ETK however says

No.Description Supplement Qty Part Number Price
01AT-Control unit Motronic HARDWARE 1.3 1 12141738168 ENDED
01AT-Control unit Motronic HARDWARE 1.3 1 12141748258 $813.86

Whatever the actual revision used, its pretty clear we have a v1.3 Motronic.

Wiki (

Motronic

) says:

DME / ECU 1.3 Info.

Tuning is set in 2D map.

Motronic 1.3 from 1989: The fault code memory is extended to contain all fault codes that are detected by the EMS.
In vehicles sold in the USA, when the ECU detects that a some faults are present it earths pin 15 and the Check Engine warning lamp on the dash will light. The lamp will stay lit until the fault is no longer present. A warning lamp is not fitted to vehicles sold in other markets. The faults that will turn on the lamp are mainly those concerned with emissions. Other faults are logged by the ECU but the lamp will remain out.

ECU (DME) sensors run off 5v

Download extended info here – http://www.opel-scanner.com/files/DME_1.1_1.3.pdf

DME bootup: (roughly paraphrased)
Interestingly enough I got to learn this properly via the tips and tricks for checking why the car doesn’t start..

Ignition switched on:
ECU earths 85 , 36 which causes relay to close and 30 is connected to 87 starting the fuel pump. (30+87 can be bridged manually under the green relay to test pump). Its opened for +- 1 second.

Next up it reads values from the CAS (Pulse Sensor). If it reads a speed signal (aka is it turning), so fuel pump stays on.

CAS is used to time the pulses (read from each turn of the flywheel), this is checked against the map file for correct speeds (need to use a multiplier for the map values vs the flywheel pins).

The Motronic ECU contains a fuel map with an injector opening time for basic conditions of speed and load. Information is then gathered from engine sensors such as the AFS, CAS, CTS, and TS. As a result of this information, the ECU will look-up the correct injector pulse duration right across the engine rpm, load and temperature range.

The injectors are arranged in two banks with injectors 1 and 3 (4 cylinder) or 1, 3 and 5 (6 cylinder comprising one bank, and injectors 2 and 4 (4 cylinder) or 2, 4 and 6 (6 cylinder) making up the other bank. Each bank is connected to the ECU via an independent ECU pin.

The Motronic 1.1 & 1.3 multi-point injection system pulses the injectors semi-sequentially and once every two engine revolutions. During engine start-up below 600 rpm the ECU pulses all injectors simultaneously. Once 600 rpm has been attained and if the ECU has received a signal from the CID sensor, each injector bank will be pulsed alternatively according to which pair of cylinders are approaching TDC. If a signal is not received from the CID sensor the injectors will remain on simultaneous operation. However, if the CID sensor subsequently sends a signal to the ECU after the engine has commenced running, the ECU will pulse the injectors semi-sequentially after the next deceleration phase – even if the CID sensor then ceases to send a signal.

—

Motronic troubleshooting (aka code readout)

The fault codes for these systems will appear on the check engine light when you turn the ignition key to the On position (but dont start it), then: fully depress the accelerator pedal, then fully release it. Repeat the depress,release cycle 5 times fairly quickly, but not too fast. It should be completed in a few seconds. What you are doing here is clicking the wide open throttle switch then the idle switch each 5 times. This signals the Motronic unit to send the fault codes by flashing the Check Engine light. You should see the light flash once then it will start flashing the 4 digit codes. If there are no faults, you should see the 1444 code (and 2444 if you have a 12 cylinder). The codes appear as a series of flashes for each digit. The flashes indicating one digit are about 1 second apart, the next digit appears after a couple second interval.

Get the codes from: http://www.unofficialbmw.com/repair_faqs/motronic.html

Other notes:

CPU most likely 8051
EEPROM 26C256, so 32KB / (32768 byte) chip. 28 pin.
Can also use a 64KB chip, just ground the relevant pins for which 32kb to use or burn one rom at appropriate location. Useful for running 2 tunes – eg original and second.

Tools available for this eg http://www.moates.net/font-size-12ti….html?cPath=31

Should run from a AT29C256 or similar.

ECU interesting pinouts:

Pin 55 TX
Pin 13 RX
Pin 2/19/24/26/14 GND

File layout (unchecked)

AFR idle: 0x662E (pos 26158)
AFR partial: 0x6662

AFR full: 0x6640

ign idle: 679A
ign partial: 0x67FE
ign full: 0x67CC

Sample bin file here: ftp://anonymous@bmw-wiki.org/ROMs/Mo…y-Motronic.bin

(I’m having issues downloading that from China, so haven’t checked it yet against the alleged layout).

Table calculations:

fuel maps correspond to (value/128 +1)*lambda_1

Bosch-Motronic f(n) = (n-30) x 0.75
some Bosch f(n) = (n-509) x 0.75

Bosch KE-Motronic und Mono-Motronic
f(n) = [(255-n)/6] +1

Motronic info

http://motronic.ws/

Very useful info for looking this up from http://motronic.ws/map201.htm

Maps 20 bytes each

Other thoughts – now we know what pinouts / values are expected in the system, guess just need to see what equipment provides appropriate inputs. eg, replace expensive spares with cheaper/better modern versions.

Most of the differences are usually in the mounting afaik, not the tech.

Motronic pinouts and board details here – http://www.e34.de/tips_tricks/motronic/m1_3.pdf

Lots of people working on this kind of thing with software like tunerpro / xdf (map) files.

*Good* forum for info (although not relevant to the Bosch ECU’s, but concepts / strategies same)
http://www.thirdgen.org/techboard/di…uide-book.html

The motronic 1.3 software in most of the E32’s should be fairly easy to modify – its all been documented well online before, so just a matter of getting a few rom files for specific models, and comparing tables.

Shogun has at least one rom (as he’s burning and selling them according to another post).
I can extract the current one from my DME and make a bin file for comparison.
We also have the rom from ftp://anonymous@bmw-wiki.org/ROMs/Mo…y-Motronic.bin

Should be enough to see where the changes are for the E32 / 8051CPU Motronic 1.3 anyway, so that I can tweak.

Another lucky thing is that I also used to write embedded software on 8051 cpu years ago (although I’ve forgotten it all by now)

Any interest in this from others (I’ll need more rom’s though)?

As my car currently has only 1 key, I thought I’d do a little bit of research into getting another one.
BMW quoted 2000rmb, which although understandable, is a little bit expensive. That doesn’t include the remote functionality in the current key either.

A check of taobao shows that you can buy blanks for slightly less – in the 30rmb range.
Mine is a 3 button remote, rather than a 2 button remote.

Images of both below:

Yes, I realise that these are just keys, and don’t contain the remote stuff either, but as a spare, its all I really need.

There is a fly in this ointment though. All European cars post 94 are mandated to include anti-theft immobilizers, which in BMW’s case involves checking for a transponder, so a key won’t start the car.
BMW calls this system EWS – Electronic Watchdog System.
A datasheet aimed at BMW workshop staff on EWS is here BMW EWS Overview and Worksheet

The infrastructure looks something like this:

I could buy a small board to bypass the EWS (about 130rmb), but I like the idea of having the anti-theft still working.

So, how does the EWS2 do the immobilization in my car?

A little bit of research shows that my current BMW (a 96 E36 import from German) uses BMW’s EWS2 for antitheft. This appears to be based on a motorola chipset ( XC424114CFN ) for the immobilizer side, and its relatively easy to read out or reflash the existing data from the chip if necessary should I need to do so.
eg if I lost all my keys, and needed to start from scratch. There are 3rd party tools galore that go into that eg CarProg, so not much point looking at that for me.

Lets look at the transponder side.
The transponder in my current key is a Philips PCF7930

This comes in a series – PCF7930, PCF7931, and PCF7935, and PCF7936

The main differences between the transponders are as follows:

PCF7930 – can read / write data.
PCF7931 – write once (one time programmable), read data.
PCF7935 – it has 2 area’s of memory general, and shadow memory, and both are read / write.
PCF7936 – it adds a crypto mode to the 7935 functionality.

The PCF7936 is also referred to as a Hitag2.
This is also the transponder current generation of BMW’s are using eg the E90/E92/E94’s

They’re quite simple devices really, a datasheet for them is here – PCF7930 / 31 / 35 Datasheet

They essentially have a few bytes of memory to work with.
32 bytes for control, and 80 bytes for user data.

Physically they’re about half the size of a fingernail and a few mm thick.

They look something like this –

A quick search on taobao shows that a PCF793x sells for between 17-20rmb or so.

So, so far I can get the key for about 30 odd, and the transponder for about 20rmb or so.

I still need to be able to program the transponder though, so whats available for that?

Another quick google shows that there is a common windows tool called anahtar which works with quite a few programmers.

As you can see, it supports quite a few pieces of hardware. Anahtar does need some hardware to talk to the transponder though, so I also need a transponder programmer.

A search for a normal RFID programmer is pretty polluted with car remote programmer spam sites.

That said, the usual result for BMW’s is the AK90 programmer. This is a bit on the expensive side – its around 1500rmb, and I really only need to do this once, so lets look for other options.

Ideally we’re looking for a low frequency rfid programmer.
If possible, I’d like to have something that does other chips too, so something that can cope with the below may be handy at some point, if only so I can steal peoples cars play with it.

RWProg looks interesting, as it has a lot of support for other rfid chips – http://www.bicotech.com/?page=prod_rwprog&lg=en.

Unfortunately a search for that on Taobao shows no results. RFID reader’s on the other hand are dirt cheap, with the cheapest usb ones going for 35rmb or so.

The specs for the PCF793x series don’t actually say what frequencies it runs off unfortunately, so its a little hard to find an appropriate device quickly.

I do note that there appear to be a disproportionate amount of card writers advertised on Taobao which claim to do Mifare, which not co-incidentally is similar enough to what is being used on the metro here in China in most cities for travel cards. I guess that means there probably are lots of fake ones around…

If I check whats usually used for programming the PCF793x series, Philips (NXP) pushes their PCF7991, while Philips doesn’t write what frequency that runs on either, this chinese site says 125khz http://www.docin.com/p-74627587.html

So, its a 125khz programmer.

Unfortunately those are 500bucks on tabao, so I keep looking.

…and Bingo, I can find a BMW key programmer which will do it for 350rmb.
Sounds good.

350 for the transponder reader/writer, 20 odd for the transponder, and 30 odd for a key.
400RMB total, and new copies of the keys will cost about 50rmb each vs the 2000rmb bmw wants.

Its a win!

Next up, what frequency does the remote use, so I can get a replacement for that…

My remote frequency is 433.92Mhz, as I have a Europe car.
I’ll guess it would probably be 315MHz if it was a US, CN or SA built car.

As I’ve had some negative comments about this all being possible, added some photos of the tool I bought to successfully clone my keys.

The box full of goodies.

Box contents

Key Cloning tool

Key Cloning tool side view

Key Cloning tool MCU adaptor side view

Box contents

Cloned keys (original on far left)

Software Side:

Installing hardware drivers
Device is a standard FTDI usb serial chip based product, needs 2 drivers installed, the first for the serial chip, then a second for the device driven by the usb to serial chip.

Installing software

Software running

Had a client ask about some .cn stuff over Skype, so proceeded to give him the whole this is what is needed now haha fun story
i.e but no, theres more, you also need this…. and this…. and this… and don’t forget the photo in front of the sign..

While we do have our own list of requirements in this blog, I did google to see if there was something less dry, and discovered a little blog post which mirrored our own experiences quite similarly. Its a good read too –

Read this http://www.candisgroup.com/blog/busdev/the-history-of-chinese-icp-and-cn-ownership-2, then follow on below, as this was intended to be posted as a comment on that post, but their blog comments appear to be borked at the moment.

I’ve fleshed it out a little, and it still seems semi relevant here.

Do remember that while anything I say can and will be used against me as evidence, I never said it. Got that? Good, now read on.
—

Glad to see I wasn’t the only one that had to go through all that, and again in October last year.

I also fondly remember the shut down their site stuff.

First we made a ICP warning page, then a blank page, and then I got creative and started mapping the IP for the domain to the miibeian.gov.cn server – you never know, maybe they’ll shut down their site!

Apparently although this was amusing, it also wasn’t enough, and we too had to vanish the DNS into nothingness NOW
(which is also hard, as explaining that things take 1-2 days to propagate doesn’t seem to sink in with the telecom people)
Mind you, the threat from the terrori^H Telecom people to shut off their site didn’t really have the same effect when its already shut off dammit!

Then there are the clients who ask why do we have 4 icp addresses?
Well, because the government decided that you could only have 1 per login when we did them in 2005. However client just has 1 site, and they point .com .com.cn .cn et al to the same place. and Telecom INSISTS that you have an ICP for all of them, even though its the same site. Luckily they seem to have fixed that now, although we still have the odd site with 3 or 4 icp’s just because you can’t seem to get them to accept they’re all the same site dammit, just let us register them to the one ICP, please!

…and then there are the mysteriously disappearing ICP numbers of doooom, where the telecom people says that ICP is invalid. Its fake I tell you, FAKE!
Although you could swear that the ICP is real as you did all that darn paperwork the last time around, plus you have an email from miibeian which has that number, and domain, so please please recheck… but they still say thats impossible, it doesn’t exist in their system.
Why, why, why, would we possibly have a bazs.cert for the domain together with the icp and the url tucked away in our files if we didn’t receive it from them?, but no, we have to do it all again because their system is buggy, and these aren’t the droi^Hnumber’s you’re looking for.

At least they seem to have dropped that requirement now for the bazs.cert file to be uploaded to all sites now. I shouldn’t say that too loud though, someone up high will probably smack their head now, and say, I knew there was some other hoop we forgot to get people to jump through now, erm, I mean new legislation thats not clearly thought out so people have to work out the hard way what we want, er no, thats not right either, forget what I said…

Or the good old China optimistic scheduling / planning with regards to implementation dates?

i.e. You have 2 weeks to get all of your 700+ client registrations completed in triplicate, filed with CNNIC, registrar, and local telecom authority or we shut their site off…. What do you mean you need some kind of notice.
Jump NOW! We say how high.

Ah, the good old days.

At least the clients are understanding when you tell them hey, the government changed the rules again, we need more stuff, come for photos, bring your id’s, and business licences, and do it asap, or you lose the domain.

It’s at time like this when I think back to what should have been the slogan for the Shanghai expo and the thing with 5 circles up in Beijing (just prior to the big bonfire and marshmallows thing at the CCTV building that I didn’t get invited to).

China. – We invented bureaucracy.

Its been a while since I did any IPCam stuff, but I’ve now got most of the bits I needed together again, as well as a new laptop for dev work (curse the thieves that stole my last one!)

As we recall from previous work, the main binary for the IPCam runs off a file called “camera”.

As some people have discovered, it likes to reboot the equipment when its not happy (eg when the camera is unplugged, it has issues talking to hardware, or when someone has flashed the wrong firmware).

So, lets take a look at the executable to see what interesting bits we can find out from it.

#file camera tells us – BINFLT file format. Fileflags: RAM GZIP.
So we know its a compressed bflt elf – bflt stands for binary flat file, and it uses gzip compression. It also sits in ram.

A hex dump of camera shows this for the first few bytes:

62 46 4C 54 00 00 00 04 | bFLT . . . .

[Note – I had about 4 pages of #$%# work done on this, and WordPress decided to flake once finished due to an errant pasted 0x0 null byte above, cutting off the rest of my post, so this is going to be shorter and angrier than it was originally written.
Lesson learned, always save stuff elsewhere before posting.]

bFLT headers consist of 4 bytes identifier, then 4 bytes for the version number.
In this case, its a version 4 bFLT file.

If we take a look at the header file source for bflt at the uclinux site we see the below layout.


struct flat_hdr {
char magic[4];
unsigned long rev; /* version */
unsigned long entry; /* Offset of first executable instruction with text segment from beginning of file */
unsigned long data_start; /* Offset of data segment from beginning of file */
unsigned long data_end; /* Offset of end of data segment from beginning of file */
unsigned long bss_end; /* Offset of end of bss segment from beginning of file */
/* (It is assumed that data_end through bss_end forms the bss segment.) */
unsigned long stack_size; /* Size of stack, in bytes */
unsigned long reloc_start; /* Offset of relocation records from beginning of file */
unsigned long reloc_count; /* Number of relocation records */
unsigned long flags;
unsigned long filler[6]; /* Reserved, set to zero */
};


It doesn’t match up properly, as the sizes or code don’t make sense (yet).

If we take a closer look, the header file has this to say:


#define FLAT_FLAG_RAM 0x0001 /* load program entirely into RAM */
#define FLAT_FLAG_GOTPIC 0x0002 /* program is PIC with GOT */
#define FLAT_FLAG_GZIP 0x0004 /* all but the header is compressed */


Ahah!

So, all but the header is compressed for a version 4 file.

Lets check this out, and see if its correct.
Excluding the initial file identifier (bFLT), our header consists of 10 longs. Thats 40 bytes long.
Lets jump to offset 40 in the file, and see what we have there.

1F 8B 08

Those of you familiar with gzipped files will recognize that – its the gzip header identifier. So, so far, so good.
Compressed files aren’t very useful to us, as they don’t show much text content.
So, we could unzip the file to take a look at whats inside.

There are a number of ways we can unzip this (zcat, gzip -d etc), but I’m going to be lazy, and use someone elses premade code.

See below for some perl to safely uncompress our binary , taken from here – http://www.openwiz.org/wiki/BWFWTools_Release


#!/usr/bin/perl

foreach my $bfltZ (@ARGV) {
expand_blftZ($bfltZ)
}


If we run that on our ‘camera’ executable, we should have an uncompressed bFLT file as output ‘camera.unz’.
Lets run strings on ‘camera.unz’, to see what interesting text content is in there.

Some interesting things of note:

From the variables list below, looks like there is a way to turn off the LED…

led_mode
ptz_center_onstart
ptz_auto_patrol_interval
ptz_auto_patrol_type
ptz_patrol_h_rounds
ptz_patrol_v_rounds
ptz_patrol_rate
ptz_patrol_up_rate
ptz_patrol_down_rate
ptz_patrol_left_rate
ptz_patrol_right_rate

We also have a full list of the internal cgi functions now, which might prove useful…

snapshot.cgi
get_status.cgi
get_camera_params.cgi
decoder_control.cgi
camera_control.cgi
reboot.cgi
restore_factory.cgi
upgrade_firmware.cgi
upgrade_htmls.cgi
get_params.cgi
set_alias.cgi
set_datetime.cgi
set_users.cgi
set_devices.cgi
set_network.cgi
set_wifi.cgi
set_pppoe.cgi
set_upnp.cgi
set_ddns.cgi
set_ftp.cgi
set_mail.cgi
set_alarm.cgi
videostream.cgi
video.cgi
test_ftp.cgi
test_mail.cgi
set_misc.cgi
get_misc.cgi
set_p2p.cgi
get_p2p.cgi
set_forbidden.cgi
get_forbidden.cgi
set_decoder.cgi
comm_write.cgi
wifi_scan.cgi
get_wifi_scan_result.cgi
get_log.cgi
check_user.cgi
check_user2.cgi
backup_params.cgi
restore_params.cgi
erase_allparams.cgi
set_mac.cgi
do_cgi: unknown cgi

You can see that Maverick decided to fake the X-Mailer smtp header (Foxmail is a commonly used Mail Program in China).

MIME-Version: 1.0
Content-Type: multipart/mixed;
boundary=”smtp_msg_boundary”
X-Mailer: Foxmail
–smtp_msg_boundary
Content-Type: image/jpeg;
name=”%s(%s)_%c%s.jpg”
Content-Transfer-Encoding: base64
–smtp_msg_boundary–

I’m interested in why the firmware reboots on some firmwares though, so lets take a deeper look at the code.

To do so, we’ll need to decompile it.
The better equipped than me will probably use something like the nice ARM decompiler plugin for IDA-Pro called Hex-Ray. Unfortunately that costs $$$, and I’m just a hobbyist.

Luckily there is a free windows decompiler called arm2html available here.
arm2html doesn’t handle compressed bFLT files though, so you’ll need to point it at the freshly ungzipped code you got from the perl script above.

As we know that the camera executable reboots after issuing i2c errors, the first piece of decompiled code I wanted to look at was the first piece of code related to i2c:

(excerpted piece below)
02588: e1a0c00d mov ip, sp
00258c: e92dd810 stmdb sp!, {r4, fp, ip, lr, pc}
002590: e24cb004 sub fp, ip, #4 ; 0x4
002594: e24dd00c sub sp, sp, #12 ; 0xc
002598: e59f023c ldr r0, [pc, #572] ; [0027dc] "/dev/i2c0"
00259c: e3a01002 mov r1, #2 ; 0x2
0025a0: eb00c9b8 bl 034c88(c9b8)
0025a4: e1a04000 mov r4, r0
0025a8: e3540000 cmp r4, #0 ; 0x0
0025ac: aa000003 bge 0025c0(3) ; jump
0025b0: e59f0228 ldr r0, [pc, #552] ; [0027e0] "%s: can not open i2c device"
0025b4: e59f1228 ldr r1, [pc, #552] ; [0027e4] "zoom_test"
0025b8: eb00bc86 bl 0317d8(bc86)
0025bc: e91ba810 ldmdb fp, {r4, fp, sp, pc} ; return

The full piece of code essentially loops 7 times trying to open the i2c sensor to call the zoom_test code. If it fails, it calls for a reboot.
Success proceeds to setting up the camera.

We know from our boot log that my camera in this model is a Sonix288.

dvm usb cam driver 0.0.0.0 by Maverick Gao in 2006-8-12
usb.c: registered new driver dvm
dvm usb cam driver 0.1 for sonix288 by Maverick Gao in 2009-4-20
usb.c: registered new driver dvm usb cam driver for sonix288

The Sonix288 is a chipset SoC that will talk to an attached image sensor via i2c. I think that the Sonix288 is probably a standard USB 1.1/2.0 compatible UVC (USB Video Class) chipset from a bit of googling about it.

We don’t have the source for our particular camera though (its secret, much like the data sheets, grrr…).
What to do?
Linux generally uses spcaxxx (and UVC) drivers for talking to camera’s, so lets start taking a look there.

http://read.pudn.com/downloads127/sourcecode/unix_linux/539050/zc030x/zc030x_cameras.c__.htm

http://www.hackchina.com/r/54654/zc030x_i2c.c__html

Taking a look at some spcaxxx driver header files and code shows that i2c is setup by first getting the USB VID, USB PID of the hardware, then talking to the i2c device on that hardware.

So, we need to know what our USB VID and PID’s are.

Generally all devices from a given manufacturer will have a single VID as issued by the USB Forum.

If we search through the code for Sonix, it appears that Sonix’s VID is 0x0c45

{USB_DEVICE(0x0c45, 0x607c)}, /* Sonix sn9c102p Hv7131R */
(from http://linux.downloadatoz.com/linux-kernel-webcams-driver-gspca-spca5xx/ )

The Linux UVC page confirms this http://www.ideasonboard.org/uvc/.

(Listings of webcams excerpted – note the VID of 0c45)

0c45:6310 USB 2.0 Camera (Trust Chat Webcam) Sonix Technology
0c45:63e0 Sonix Integrated Webcam (Dell notebooks) Sonix Technology
0c45:63ea Laptop Integrated Webcam 2M (Dell Studio 1555 notebooks) Sonix Technology
0c45:6409 USB 2.0 Camera (Nokia Booklet 3G netbooks) Sonix Technology
0c45:6415 Laptop Integrated Webcam 1.3M (Dell Inspiron 13z notebooks) Sonix Technology

Ideally at this point, I’d have a data sheet for the Sonix288 chip, but the #$%#$ people at Sonix don’t seem to publish one for us mere mortals.

So, we’ll use the next best thing, and use one for their other chipsets.

SN9C1xx PC Camera Controllers –
http://ww2.cs.fsu.edu/~rosentha/linux/2.6.26.5/docs/video4linux/sn9c102.txt

There is a lot of good useful info in that particular file. We don’t know how much is useful yet, but generally chipsets are quite similar for a given range.

Image sensor / SN9C1xx bridge | SN9C10[12] SN9C103 SN9C105 SN9C120
——————————————————————————-
HV7131D Hynix Semiconductor | Yes No No No
HV7131R Hynix Semiconductor | No Yes Yes Yes
MI-0343 Micron Technology | Yes No No No
MI-0360 Micron Technology | No Yes Yes Yes
OV7630 OmniVision Technologies | Yes Yes Yes Yes
OV7660 OmniVision Technologies | No No Yes Yes
PAS106B PixArt Imaging | Yes No No No
PAS202B PixArt Imaging | Yes Yes No No
TAS5110C1B Taiwan Advanced Sensor | Yes No No No
TAS5110D Taiwan Advanced Sensor | Yes No No No
TAS5130D1B Taiwan Advanced Sensor | Yes No No No

Interesting… Hmm, so it seems that Sonix uses a SN9Cxxx for its product names.
Lets google for SN9C288 instead, and see if we get any results.

Bingo.

From here – www.lh-invest.com/en/showpro.asp?id=308&proid=3

* MCU: SONIX SN9C288
* Sensor: MICRON K14 1,300,000 pixels CMOS
* 5-glass lens,can reach 30 frames/sec. under 640*480
resolutions,software insertion value 1,300,000 pixels
* Focus range: 3CM to infinitude farness
* Dynamic video resolutions: 1280*960 pixels(max.)
* Support microsoft UVC driver-free function
* Support RGB24 and YUY2 two kinds of image formats
* USB 2.0 port,support plug and play
* Human face tracking function

Seems our chipset is finally getting some details.
Max resolution, video modes, face tracking capability, etc
We also know that it can be paired with a Micron K14 image sensor.

Further googling reveals it can also be paired with the MI-0360 (which is listed above).

SN9C288+MI0360

This page also gives us a possible pid:

http://forums.lenovo.com/t5/SL-Series-ThinkPad-Laptops/camera-problem-in-all-windows-Creation-of-the-Video-preview/m-p/163019/highlight/true

“Device Name: ?USB Video Device

PnP Device ID: VID = 0C45 PID = 62C0
Serial Number: 6&&2BCAFCF3&&0&&0000
Revision: (Information not returned)

Device Type: Standard USB device – USB2.0 High-Speed

Chip Vendor: SONiX
Chip Part-Number: SN9C288PFG

In Microsoft parlance, this looks like this USB\VID_0C45&PID_62C0

Googling that gives us a product with windows drivers (HP) and more.

http://www.downloadwindowsdrivers.info/usb/vid_0c45/pid_62c0/mi_00/

Also says that these product drivers also work.
USB\VID_0C45&PID_62C0 ;SN9C211/213/230

That means we can take a look at their inf file and see if anything useful in there. Unfortunately I did, and there isn’t much 🙁

Sonix has datasheets for some of their other products available though (SN9C2028AF).

First, a quick overview of UVC devices (snarfed from elsewhere).

USB devices are required by the USB specification to respond to the Host device (the computer) with a stream of data describing the device and the interface to the device. This “Device Descriptor” includes vendor, product, and version IDs specific to the manufacturer, the product, and the version of the product. In addition to the device information, the descriptor also includes information on how to talk to the device through a series of “Interface Descriptors”.

The device descriptor for Video is 13.

#define USB_DEVICE_CLASS_VIDEO 0x0E

In addition to the end points described in the interface descriptors, all USB devices support a control pipe to end point 0. This is used to manipulate some of the low level functions of the device such as power, and error status queries.

The USB Video specification describes two interfaces, a control interface to manage the camera, and a stream interface to send or receive video information from the camera.

The control interface is used to manipulate the camera parameters such as brightness and contrast as well as to negotiate a valid set of the video format, frame size, frame rate, and compression rates parameters that describe a video stream. In addition, the control interface can ask the camera for still frame.

In the datasheet for the 2028, it basically states that they use end point 0 for STD Commands, with a maximum packet size of 64 bytes.

Further googling reveals that there is no special driver for it, its a plain UVC 1.0 device.

usb 1-2: New USB device strings: Mfr=2, Product=1, SerialNumber=0
usb 1-2: Product: USB 2.0 Camera
usb 1-2: Manufacturer: Sonix Technology Co., Ltd.
Linux video capture interface: v2.00
uvcvideo: Found UVC 1.00 device USB 2.0 Camera

Further heavy baidu’ing in Chinese sites finds that the SN9C213 and SN9C288 are the same pretty much.

其内部编号是SN9C213，功能完全和SN9C288一样。From http://ep.cbifamily.com/2007/04/44/87819.html

David McCullough very nicely also compiled in usb debug support on his kernel and ran some tests too:

> > > T: Bus=01 Lev=01 Prnt=01 Port=00 Cnt=01 Dev#= 2 Spd=12 MxCh= 0
> > > D: Ver= 2.00 Cls=ef(unk. ) Sub=02 Prot=01 MxPS=64 #Cfgs= 1
> > > P: Vendor=0c45 ProdID=62f1 Rev= 1.00
> > > S: Manufacturer=Sonix Technology Co., Ltd.
> > > S: Product=USB 2.0 Camera
> > > C:* #Ifs= 2 Cfg#= 1 Atr=80 MxPwr=500mA
> > > I: If#= 0 Alt= 0 #EPs= 1 Cls=0e(unk. ) Sub=01 Prot=00 Driver=(none)
> > > E: Ad=83(I) Atr=03(Int.) MxPS= 16 Ivl=6ms
> > > I: If#= 1 Alt= 0 #EPs= 0 Cls=0e(unk. ) Sub=02 Prot=00 Driver=(none)
> > > I: If#= 1 Alt= 1 #EPs= 1 Cls=0e(unk. ) Sub=02 Prot=00 Driver=(none)
> > > E: Ad=81(I) Atr=05(Isoc) MxPS= 128 Ivl=1ms
> > > I: If#= 1 Alt= 2 #EPs= 1 Cls=0e(unk. ) Sub=02 Prot=00 Driver=(none)
> > > E: Ad=81(I) Atr=05(Isoc) MxPS= 256 Ivl=1ms
> > > I: If#= 1 Alt= 3 #EPs= 1 Cls=0e(unk. ) Sub=02 Prot=00 Driver=(none)
> > > E: Ad=81(I) Atr=05(Isoc) MxPS= 512 Ivl=1ms
> > > I: If#= 1 Alt= 4 #EPs= 1 Cls=0e(unk. ) Sub=02 Prot=00 Driver=(none)
> > > E: Ad=81(I) Atr=05(Isoc) MxPS= 600 Ivl=1ms
> > > I: If#= 1 Alt= 5 #EPs= 1 Cls=0e(unk. ) Sub=02 Prot=00 Driver=(none)
> > > E: Ad=81(I) Atr=05(Isoc) MxPS= 800 Ivl=1ms
> > > I: If#= 1 Alt= 6 #EPs= 1 Cls=0e(unk. ) Sub=02 Prot=00 Driver=(none)
> > > E: Ad=81(I) Atr=05(Isoc) MxPS= 956 Ivl=1ms

The uvc driver status is obviously an issue as we’re on a 2.4 Kernel. No uvc support on 2.4.
So…, we either compile 2.6 with UVC support, and reflash, or we continue to use Mavericks driver in lieu of any source.

With that, you have some background on things..
Now, why does this cause a reboot on some machines?

Well, the hardware is different, so the hardware isn’t seen by the driver.
From looking at a few different boards I have seen a few devices id’s used.

So far I have seen these:
vid_0c45/pid_62f1 – (The Sonix Chipset allows you to write pid’s into the device, so this can be changed, or alternately change the driver from 62c0 to 62f1 on these models)

vid_0c45/pid_62c0 – (Our driver is compiled for this)

For those of you with rebooting machines, remove ‘camera &’ from the boot sequence, recompile the kernel with USB verbose debug message logging, and start posting your vid/pid’s here so we can compare, and add to the list.

If someone twisted my arm I could probably oblige…

References:
http://www.beyondlogic.org/uClinux/bflt.htm – bFLT file format details.
http://www.garykessler.net/library/file_sigs.html – Common file format headers
http://www.openwiz.org/wiki/BWFWTools_Release – bFLT unzip and other tools
http://www.hex-rays.com – IDA Pro and Hex-Ray ARM Decompiler
http://www.sigmaplayer.com/filebase.php?d=1&id=13&c_old=5&what=c&page=1 – Arm Decompiler
http://www.ideasonboard.org/uvc/
http://read.pudn.com/downloads127/sourcecode/unix_linux/539050/zc030x/zc030x_cameras.c__.htm
http://www.hackchina.com/r/54654/zc030x_i2c.c__html
http://linux.downloadatoz.com/linux-kernel-webcams-driver-gspca-spca5xx/
http://ww2.cs.fsu.edu/~rosentha/linux/2.6.26.5/docs/video4linux/sn9c102.txt

Files from this post:
(arm2html.exe, bFLT gunzip perl script, original camera bFLT, uncompressed camera bFLT, and camera asm source )
Camera Disassembly, unpacked bFLT and Tools

As I’ve been doing a bit more hardware stuff recently, I thought I’d get some more tools.

Yes, Taobao is wonderful 🙂

I already have a scope – the eminently hackable Rigol DS1052e 50mhz (now running at 100mhz cough).
I already have a bench PSU.
I also have a JTAG device (although its a chinese clone, but hey, it works).
… and I also have a whole bunch of ttl -> rs232 adaptors (as they fry when you’re not careful).
I also have a 3d printer 🙂
I also have a few Arduino bits and bobs, as well as some much more capable ARM dev kits (which I prefer).

What more could I want?

Well, a bus-pirate – but Seeeeeeeeeeed studio still haven’t gotten their sh*t together, and I can’t seem to order from them.
Grumble. Tried again today, but my credit card is rejected as usual. Works on every other site, but theirs..
Still, whilst checking out Seeed Studio’s blog I noted that they mentioned Shanghai (finally!).

Apparently, there’s a hardware hacker dev lab *right* by where I used to live.
They’re up on 50 Yong Jia Lu / website is – http://xinchejian.com/.
I still can’t believe I hadn’t heard about them.

Also slightly annoyed that I missed the last few open days they had.

Going to have to go visit after CNY, and make some friends there 🙂

Lawrence.

The new year has arrived, and so has cheaper pricing for internet access at home.

Currently for home use, there are 3 options for internet over fixed lines.

• China Telecom with ADSL over Copper / FTTB+Lan / FTTH
• China Unicom
• Orient Cable

Incumbent China Telecom has reduced their rates significantly since last year, as Unicom is encroaching on their space.

Lets start with Unicom’s packages:
The latest pricing for 2011 for Unicom is here:

http://www.sh.chinaunicom.com/family/ywcp/jtkh/kdl/index.html#lt_dw_md

Unicom can provide up to 30M for home use (assuming your area can have Fibre access via FTTB+Lan or FTTH)

10M is 198 /month
20M is 248 / month
30M is 288 / month

If you don’t have fibre, and can only get ADSL lines, then your max speeds will be 4M – 6M depending on distance from the local substation.

4M is 144 / month
6M is 168 / month.

Shanghai Telecom has committed to rolling out Fibre to all users by 2013 though, so most areas will start to see fibre availability coming soon.

Shanghai Telecom

Shanghai Telecom pricing is obviously being directly targeted by Unicom’s. Each Unicom price point has been aimed squarely at beating Telecom’s.. Competition is good, although Unicom could do better. Shanghai Telecom has far better backend infrastructure though, and that’s going to take time for Unicom to improve on.

Shanghai Telecom’s current best value package is this:

http://sh.ct10000.com/pptc/ehome/bundle/e8/gwe8sqb/

This offers 10M internet (again with the caveat of Fibre availability in your area) for the sum of 188RMB a month. If you don’t have Fibre, they reduce that price by 10RMB and provide the standard 2M / month (which in comparison with standard adsl rates is not a good deal).

They sell this as a bundle with Telephone access also, so you get reduced phone rates too.
This includes a few other random things like incoming caller ID and custom ringtones for your callers, as well as 30hrs of monthly wifi access assuming you need to use or can find their wifi when around town.

Bundles are:

E8 – ( http://sh.ct10000.com/pptc/ehome/bundle/e8/gwe8sqb/ )
10M – 188rmb /month.

E9 (http://sh.ct10000.com/pptc/ehome/bundle/e9/zxe92011/)
20M – 369rmb / month

Orient Cable

Last and least, we have Orient Cable. OCN rates are here
http://www.ocn.net.cn/gsgg_cuxiao07.html

I’m not going to bother listing their package in detail, as its pretty crappy.
1M for 110 / month.

Looking at the options, my pick for best choices would be

Fibre users:
China Telecom at 188rmb / month with 10M if you can get it value for money wise.
For the speed demons China Unicom for 30M / month.

ADSL only:
China Unicom for their 4M or 6M options if you can only get ADSL installed in your area.
Unicom’s may not offer this in your area though, so you may be stuck on the standard 2M for 150/month till they upgrade lines.

Good luck!

China Mobile may start coming to their senses at some point and offer unlimited fixed wireless, but for now their offerings are too expensive for home use.

Lawrence.

One or two of our servers have been a little bit overloaded recently.

They’re going to be replaced with beefier machines, but due to a number of issues I haven’t been able to replace them yet.

Issue #1 – Pre expo, we weren’t allowed to replace anything.
Issue #2 – Post Expo, I’m no longer allowed in the data center!

We’re working on sorting issue #2 out, but in the interim I need to keep the older machines running.

I was previously using Monit to monitor system load.

Monit would be a good solution – it has a web ui, it can stop services if system load goes too high, and generally works when everything else is failed. This is great when things go poopy, but it has one fatal issue.

It doesn’t know how to restart stuff if load is back to normal.
This typically means that something will put the server load into unusability for a sustained period of time (due to lots of visitors), monit will go ooh, apache has gone awol, and stop it.
Unfortunately if its back to normal, monit doesn’t have a way to start it up again, so I need to manually go to the monit page, and start the service. I do get emailed on things like this, but it leads to complaints from the 2 clients that appear to monitor their particular websites more than monit does.

So, I’ve been looking at other solutions.

One such solution is sysfence

While sysfence is severly underdocumented, the examples provided don’t even work!, and appears to be abandoned, it does do the job.
Sysfence is a no bells and no whistles precursor to monit, but it has that killer feature that monit is missing.

So, how do we use sysfence?

apt-get install sysfence

Will install it, but unfortunately no config is installed.

So, start off by creating a /etc/sysfence folder

mkdir /etc/sysfence
cd /etc/sysfence

We’ll need to create a config file for it, so

pico sysfence.conf

My sample sysfence script is below (explanation underneath script)


rule "ApacheStop" {
la1 >= 10.00 or la5 >= 6.0
}
run '/etc/init.d/apache2 stop;'


rule "ApacheStart" {
la1 <=2 } run once '/etc/init.d/apache2 start;'

rule "warning" { la1 >= 8.00 } run once 'echo "Load High: BACKUP" | mail lawrence@computersolutions.cn'


I'm having issues with apache causing load to rocket, so I've setup some rules as follows:

If load average for 1 minute > 10 (ie server is going bonkers), and load average for the last 5 minutes > 6 then stop apache.
if load average for 1 minute > 8 send me an email.
if load average for 1 minute < 2 then start apache. This will only run one time if load is below 2. The documentation http://sysfence.sourceforge.net/ goes over how to write a rule. Note that the examples are broken;

eg
if {
la1 >= 8.00
} run once 'echo "SHOW FULL PROCESSLIST" | mysql | mail my@email.com'


Issue? All rules need to have a "rule name" specified.

So a corrected working version would be:

if "some rule" {
la1 >= 8.00
} run once 'echo "SHOW FULL PROCESSLIST" | mysql | mail my@email.com'


Back to our setup..

Now we've setup a ruleset, we need to run it. Calling sysfence /etc/sysfence.conf

Will run it as a daemon.

ps -ef shows our rulesets running:

root 7260 1 0 05:51 ? 00:00:01 sffetch
root 7261 7260 0 05:51 ? 00:00:00 sfwatch 'warning'
root 7262 7260 0 05:51 ? 00:00:00 sfwatch 'ApacheStop'
root 7263 7260 0 05:51 ? 00:00:00 sfwatch 'ApacheStart'

sffetch is the daemon, and sfwatch are the rules it runs.

As sysfence is quite rudimentary, you'll need to kill it if you change rules.

You'll also need to add it to your startup scripts or create one. I'll be lazy and not go over that right now. If people are interested, add a comment, and I'll put something up.

Sysfence can be downloaded here - http://sysfence.sourceforge.net/ (or via apt-get if on a Debian based OS)

Man page for sysfence below (note examples require adding "rulename" after if... { or rule ... {):

NAME

SYNOPSIS

DESCRIPTION

USAGE

EXAMPLES

AUTHOR

Comment spam is a big load on servers – recently troubleshooting some intermittent load issues on one of our webservers, I discovered that one page was getting hammered by comment spam abuse – the page had already generated over 400,000 comments, which was causing the server to be slightly unhappy loadwise.

After clearing out the comment table for that clients site, I looked into solutions.

mod_security was an obvious one, and the atomic corp rules seem to be better than the default mod_security ones (which break most popular apps, sigh).  However, although the gotroot rules were good at blocking comment spam, they don’t block the ip’s, so persistent spammer bots will still hammer the server.

So, whats a solution?
Blocking the ip’s dynamically for a short period.
Assuming you have a standard setup, the below should be of interest.

I stick my ruleset into /etc/modsecurity2, so amend your url’s accordingly.

My default initialise code is below (pretty much stock except for the end part)

/etc/modsecurity2/modsecurity_crs_10_config.conf

SecRuleEngine On

 SecRequestBodyAccess On

 SecResponseBodyAccess On

 SecResponseBodyMimeType (null) text/html text/plain text/xml

 SecResponseBodyLimit 2621440

 SecServerSignature Apache

 SecComponentSignature 200911012341

 SecUploadDir /var/asl/data/suspicious

 SecUploadKeepFiles Off

 SecAuditEngine RelevantOnly

 SecAuditLogRelevantStatus "^(?:5|4(?!04))"

 SecAuditLogType Concurrent

 SecAuditLog /var/log/apache2/modsec_audit.log

 SecAuditLogParts ABIFHZ

 SecArgumentSeparator "&"

 SecCookieFormat 0

 SecRequestBodyInMemoryLimit 131072

 SecDataDir /var/asl/data/msa

 SecTmpDir /tmp

 SecAuditLogStorageDir /var/asl/data/audit

 SecResponseBodyLimitAction ProcessPartial

SecPcreMatchLimit 100000

SecPcreMatchLimitRecursion 100000

SecDebugLog /var/log/apache2/modsec_debug.log

SecDebugLogLevel 3

### Ruleset additions for blocking ########################################

# Make sure to clear the default action

SecDefaultAction phase:1,pass

# Initialize collection and deprecate by 3 points per day (86400 seconds)

SecAction phase:1,initcol:IP=%{REMOTE_ADDR},deprecatevar:IP.spam=3/86400,nolog

# If there are already >15 spam points for this IP, then drop

# the connection and add 1 point (instead of 3, as below).

SecRule IP:spam "@gt 15" phase:1,setvar:IP.spam=+1,drop,setenv:spam=spam

# Clear the default action for any mod_security rules later in httpd.conf.

SecDefaultAction phase:1,pass

### End Ruleset additions for blocking ####################################

What does it do?  Basically we keep score of any ip address that visits the site.
Every day any scores depreciate by 3 points per ip.
If any ip scores more than 15 points, we block it.

Now, as it stands, its pretty useless, as we don’t do any scoring.

So, next we’ll need to add some scoring to some of the rules.

In our case, we were getting seriously hammered by pharmacy spam (mostly out of latvia)

So, I went into the default gotroot ruleset that I installed into /etc/modsecurity2

Comment spam is in this file: 30_asl_antispam.conf

I added some scoring there.
In my case, we saw hundreds of thousands of attempts for levitra and tramadol spam, so I went to the pharmacy scoring, and added this:

setvar:IP.spam=+15 in the SecRule check.
This means that any positive lookup for that rule will give the ip a 15 point score.
That will block that attacker ip from the site for 5 days (as we decrement 3 points per day for a given ip).

The setvar needs to go after the “capture,…” part of the rule.

eg

(don’t copy, paste this, you’ll need to add “setvar:IP.spam=+15” into your existing file)
# Rule 300040:
SecRule ARGS|!ARGS:/domain/|!ARGS:description|!ARGS:redirect_to|!ARGS:setting[banemail]|!ARGS:/username/|!ARGS:/user_name/|!ARGS:/page_content/|!ARGS:/search/|!ARGS:/email/!ARGS:Mensaje|!ARGS:/product/|!ARGS:/domain/|!ARGS:description$
"capture,setvar:IP.spam=+15,id:300040,rev:7,severity:2,msg:'Atomicorp.com - FREE UNSUPPORTED DELAYED FEED - WAF Rules: Spam: Pharmacy',logdata:'%{TX.0}'"

I added this to appropriate places, and watched for our first victim^Mspammer in our site logs.
Bingo – here we have a spammer trying to add to a url on one of our sites:

[23/Dec/2010:11:39:45 +0800] [www.chou.cn/sid#bcf3f338][rid#b339b670][/gallery/papratiti/img_5993.jpg.php][1] Access denied with code 403 (phase 2). Pattern match “(?:buy[-_ ]?(cheap)?[-_ ]?(?:adipex|suboxone|pseudovent|topamax|trazodone|prevacid|zyrtec|xenical|toprol|zoloft…” at ARGS:comment. [file “/etc/modsecurity2/30_asl_antispam.conf”] [line “194”] [id “300061”] [rev “10”] [msg “Atomicorp.com – FREE UNSUPPORTED DELAYED FEED – WAF Rules: Spam: Pharmacy”] [data “xanax”] [severity “CRITICAL”]

[23/Dec/2010:11:43:16 +0800] [www.chou.cn/sid#bcf3f338][rid#bfac9d68][/gallery/papratiti/img_5993.jpg.php][1] Access denied with connection close (phase 1). Operator GT matched 15 at IP:spam. [file “/etc/modsecurity2/modsecurity_crs_10_config.conf”] [line “40”]

Bam, that ip is blocked (for 5 days).  Next time around, as its the first rule to run, it will block immediately without processing other rules.

You can amend the block times by decreasing the increments or scoring as per your requirements.

All in all, this is an easy amendment to make to the gotroot rules, and makes for faster experience for users.
Thanks to http://linux.icydog.net/apache/commentspam.php for his implementation, this is pretty much based off that.

Lawrence