[NTLK] Linear Flash Question

James Fraser wheresthatistanbul-newtontalk at yahoo.com
Thu Dec 6 11:10:57 EST 2012


I was trying to take a look at how the CIS (Card Information Structure) on a typical Linear Flash memory card works and whether the number of P/E cycles a particular card had undergone might be part of the information retained in the CIS (i.e. one of the "tuples").  From what little I could understand:


the answer is "no" (i.e. the CIS is a hard-coded, factory-programmed set of parameters for the card), which put an end to that line of thought, but then I saw this tidbit in an article on wear leveling:

>many flash devices have one block with a specially extended life of >100,000+ cycles that can be used by the Flash memory controller to track >wear and movement of data across segments. 

(As a sidebar, I've been led to understand that the MessagePad has some form of this wear leveling implemented to ensure even wear across its internal flash memory.)

The question I'm trying to get to is: is there a way, via hardware, to gain access to this block so that you could get an approximate idea of how much life is left in a card (or rather, roughly how many P/E cycles remain)?  I know that flash media eventually fails, but the idea here would be to take a large chunk of the guessing out of when a given card will fail.

The reason I ask is that, some time ago, PCBman (aka D. Humphreys) mentioned some PCMCIA testing equipment he has:


(A [somewhat] working link to the Elan Digital Systems page referenced there can be found here: http://tinyurl.com/an4cdeh   )

That is, he has the ability to test a given card, but the equipment (remarkable as it is in some respects) is limited to the low-level reading and writing to which PCBman refers in his post.  Using hardware and software, he can tell whether a given card is bad or not (i.e. whether or not it can be written to or read from). 

What I'm interested in is whether there is hardware/software that takes things a step further and makes it possible to determine approximately how much life is left in a given Linear Flash card.  Is there such a combination?  


James Fraser

More information about the NewtonTalk mailing list