I'm not sure I got fully what functionality you need, but probably the "my_XXX" files can be used for this task.
If you use an ecu file "8D0907551M.ecu", then also the file "my_8D0907551M.ecu" is read by the logger if it exists.
You could store your manually added variables in this file. Same thing for the logger config, a file "my_8D0907551M_template.cfg"
will be read by the logger if it exists.

For the checksums topic, I have to read and look up some things at first.

I did NOT get around the 64 variable limit. The driver is just a basic example, I must even have forgotten to protect against that limit in Winlog.
But if you are logging variables for graphical display, I wonder if anybody can watch 64 things simultaneously.

The Winlog driver was just the best example to show how ME7L could be integrated into other programs using the shared library.
There will be lots of possibilities to integrate ME7L into your own applications, e.g. remote control of logging and streaming
of log data over network, merging log data with results from other sources, .... I'm looking forward to see what will show up.

I would be happy to look at the code if  I could get some hints of what to look for in IDA pro.
Somehow the need to connect without engine running is even more annoying when using Winlog.

To have double slowinit work while engine is running, I had to patch three locations.
In the image 8D0907551G, 0001, 1037354123, 0261206776 these locations were
found at following fileoffsets:
02B8BC:   3D04 -> CC00    // Replace JMPR by NOP
067026:   3D08 -> CC00    // Replace JMPR by NOP
067EF8:   3D03 -> CC00    // Replace JMPR by NOP

At all three locations it is checked if nmot_w is zero:
82'B8B8: F2F47AF8       MOV     R4, [nmot_w]
82'B8BC: 3D04           JMPR    cc_NZ/NE, L_82B8C6 (82'B8C6)        ; PATCH: NOP

86'7022: F2F47AF8       MOV     R4, [nmot_w]
86'7026: 3D08           JMPR    cc_NZ/NE, L_867038 (86'7038)        ; PATCH: NOP

86'7EF4: F2F47AF8       MOV     R4, [nmot_w]
86'7EF8: 3D03           JMPR    cc_NZ/NE, L_867F00 (86'7F00)        ; PATCH: NOP

If you disassemble the 551G image, you can hopefully determine some code patterns
around the mentioned locations that you can then find also in your image.

You gave all I needed, almost too easy. 
Thank you VERY much!

I'll help those who work with 551M.


Not double checked or tested in car, but everything should be ok.

edit:
This 3 position patch worked very well, really nice. Thank you setzi62!  I've not tested the 551M but the same patch in another bin.
From what I can gather I cannot see any problem making this patch permanent.

I thought it was a cluster issue and since mbox binary is widely spread I assumed some would have connection trouble without double-slow.
Ohh well, good thing it only took 5min to find 'em. 

The  -p switch for comport selection doesn't work for me.

I get "unable to open port error" when trying to use any port other than the port 2 default.

I have no problem keeping my cable on port 2, but I thought that I would share. This particular laptop has never run any previous version of ME7L, so I can't confirm that the issue is with the update and not something else.

However, ME7Info doesn't read the my_XXX.ecu automatically? And it will not create a my_XXX-template.cfg automatically?


You mean if you specifiy "MELogger logconfig.cfg" it will automatically also read my_logconfig.cfg?

I assume the my_XXX.config doesn't need all the header stuff, just a plain list of additional variables?