Circular buffer devices?

  • Release Candidate 6
    Guest:
    We are at a “proposed final” true release candidate with nothing known remaining to be changed or fixed. For the full story, please see this page in the "Pre-Release Announcements & Feedback" forum.
    /Steve.
  • Be sure to checkout “Tips & Tricks”
    Dear Guest Visitor → Once you register and log-in:

    This forum does not automatically send notices of new content. So if, for example, you would like to be notified by mail when Steve posts an update to his blog (or of any other specific activity anywhere else), you need to tell the system what to “Watch” for you. Please checkout the “Tips & Tricks” page for details about that... and other tips!

    /Steve.

misiu_mp

New member
Nov 10, 2023
3
0
I can think of a controller that would keep first few GB of sectors addresses and the rest use as a circular buffer for incoming write requests. The stored data doesnt need to be contigous in the physical storage - the controller could store a mapping between the requested address and the actual allocated address, which is not much different from what SSDs already do if I understand things right. This way the partition and filesystem, as well as _last_ few GB of written data (independent of where it was written) will be seemingly preserved, but the rest will be continuously overwritten. I dont think validrive would be able to detect such devices, as it writes a small amount of data for speed, which the driver could very well allocate in such a circular buffer fashion. Only way to detect this would be to write and verify an amount of data that is more than the actual storage in the device. Comments?
 
You're 100% correct. What you have described would be called an "MRU" — Most Recently Used — drive, and this has been discussed extensively in GRC's newsgroups while ValiDrive was being created. It's very clear that the only way to be 100% certain of a drive's storage capacity would be to write cryptographically strong pseudo-random data to the drive's entire declared storage media, then re-read the entire drive, comparing what was read to what was written.

No one argues that. But no one has seen the theoretical drive that you have suggested. COULD it exist? Absolutely. DOES it exist? Not so far as anyone knows. And why would anyone create such a thing? The logic that the MRU drive would use would need to be designed from scratch since no existing silicon would do such a thing. Why would anyone go to that length? Instead, what we see out in the world is a tweak to existing flash memory controllers to simply mis-report the drive's size. Otherwise, they're just regular smaller drives.

ValiDrive was designed to detect the fraudulent drives we know exist, and to do so non-destructively and quickly. I've designed a solution that could perform the unspoofable drive validation without destroying the data that's there... but, again, so far no "thumb drive" has acted this way.
 
Ok, so such drive would most likely have to use one of the more advanced controllers, such as those in sata SSDs or NVMes. Maybe there is an opening for fake, Validrive-resistant SSDs, just waiting for proper economical conditions...
 
No. There is no “advanced controller” that allocates in a circular buffer fashion. Why would anyone ever design such a thing? Not only have we never seen such a thing, it would never make any sense for mass storage.

Understand that these are not advanced fraudsters. All they are doing is making a tiny configurable tweak to real controllers to say that there's far more memory than there really is. They are not design anything from the ground up.
 
Ok, so such drive would most likely have to use one of the more advanced controllers, such as those in sata SSDs or NVMes. Maybe there is an opening for fake, Validrive-resistant SSDs, just waiting for proper economical conditions...

"SM2320 external portable SSD controller solution features a USB 3.2 Gen 2 interface, advanced SSD controller architecture and four NAND channels" - goes straight to USB. It's like a USB flash drive on steroids. I think in time the difference between USB flash drives as we know them and SSDs will disappear and for a while it may be really confusing. Giving frauds the opportunity to 'market' USB flash drives as SSDs (they already do of course).