Flooded G-List in MQ04xxxyyy

[RecuperoDati299]

Hello, 
I suppose you have already come across some MQ04s with a high number of G-List entries. 

In a good number of such drives you don't end up with a clicking drive, but you cannot access the LBAs.

 PC-3000 utilities do show the G-List entry count but, as we know, they cannot wipe the G-List. 

Has anyone done a deep ROM reverse engineering, hopefully to find a command for that purpose? 

I'm asking because our impression, after many exchanges with ACELab, is that they haven't been investing in Toshiba HDD research and reverse engineering for a long time now. 

Thank you for any kind feedback

Robert @ RecuperoDati299

Italia

[pbzcbf...@gmail.com]

Can PC3000 dump the tracks?

[teofi...@gmail.com]

I have the same problem. I tried to dump the tracks out of patient but nothing comes out just errors. I'll try to find good one to try it out.

[RecuperoDati299]

Hello

the save tracks function in the acelabb MQ04 looks to be sort of a fake

it does appear a mask with odd tracks pre checked out , and I tried to select both ,

but it hangs on each track , slow process and final result ... the content of each file is a typical Acelab pattern

they substitute the not good content with hex values corresponding to the string DE AD ... "dead"

Are you already taken advantage of claude.ai for these analysis?

I gave it a try. The result looks to be that in the ROM there is nothing about G-List or S-List whilst everything looks in the media of the SA.

So it would be cool to find any ATA or UART command for dumping the SA tracks and even better , for a future step, writing them back (hopefully properly modified)

With a deep disassembly, it looks like there are over 100 commands from terminal and even more via ATA.

Also looks pretty confirmed that there will be context when a password will be asked "INPUT PASSWORD CHARACTER" , singular, and it looks like that the password is really 1 character that should be a '$'

Ottieni BlueMail per Android

[pbzcbf...@gmail.com]

I examined the track dumps for MQ01. They consist of a 0x400-byte block of random data that is repeated from the start of the .trk file to the end. The .trk files are identical. It's just junk.

The G-list file in newer Toshibas does not have a "GL" signature, and the file is truncated. In fact, I'm told it's not a real G-List, but some artefact that is retrieved from the drive's SDRAM.

If you upload an example, I'll show you what I have found.

[pbzcbf...@gmail.com]

I don't know if it helps, but module 0x01 in MQ01 is the GL file.