[Cryptography] My ignorance and drive firmware hacking

Jerry Leichter leichter at lrw.com
Thu Feb 26 11:40:04 EST 2015


On Feb 25, 2015, at 3:25 AM, Peter Gutmann <pgut001 at cs.auckland.ac.nz> wrote:
>> A couple of years ago I had a (Seagate, 1TB) disk drive in a PC "fail" .... I sent it to a company ... to recover data.... They did their job
>> nicely, and their verdict on the drive was "firmware corruption".
> It doesn't necessarily need to be rewritable to deal with a firmware
> corruption issue, it could just be a case of swapping out the controller board with one from another drive of the same model, which used to be a standard means of dealing with failed controller issues.
> 
> (Glossing over a lot of details here, e.g. whether the controller loads its firmware from the host protected area and lots of other technicalities).
I'd heard - I don't recall the source, and I don't know if it's true - that modern disks and controllers are "bonded at the hip":  There is factory equipment that determines a variety of parameters relevant to a particular physical disk and records these on the controller board.  The firmware uses these parameters to correctly access the disk.

A controller without the right set of attributes will have trouble operating a disk properly.  Unless you have access to the equivalent of the factory equipment, there is no practical way to replace a controller board (and even then, existing data on the disk may be difficult to read).

Again, I don't know if this is true - but it's consistent with a general trend in manufacturing:  Rather than keep pushing the limits and bring variances in manufactured parts ever downward, accept that individual units vary and compensate in software.  (Note that biological systems have evolved to do exactly this:  Your eyes, for example, present highly distorted and idiosyncratic information about the visual field in front of them to your brain, which corrects for the distortions to produce a useful image.  Look at the world through a set of prisms that shift everything to the right, and when you reach for something in front of you, you'll actually aim to its left.  But after a couple of minutes, your difficulty will disappear - until later, when you take the prisms off and find yourself aiming too far to the right!)
                                                        -- Jerry



More information about the cryptography mailing list