Re: RFC: detect and manage power cut on MLC NAND (linux-mtd Digest, Vol 144, Issue 70)

[Andrea Marson <andrea.marson@dave.eu>]

>> 2) IIUC Boris has worked on SLC emulation too.
> No I didn't. What makes you think I did ?
My bad. I misunderstood a sentence in your slides for ELC talk :(

>
>> This seems to be a
>> promising feature because it would allow to partition NAND flash and to
>> create higher reliability partition (at the cost of halving the size).
>> Is it possibile to implement such functionality in software stack
>> only(MTD/UBI) or is it necessary that NAND memory supports specific
>> features?
> I guess you're talking about the slides for my ELC talk.
> Note that I'm just talking about hypothetical changes, and no
> development has been started on this topic yet.
I see.


>
> IIUC, you want to select SLC/MLC mode on a per-partition basis, that
> should be feasible, and this would be particularly useful for bootloader
> partitions (where UBI is not involved).
That's exactly what I was thinking about.
However it is still not clear to me how this should be implemented. I 
have read a datasheet saying that "SLC mode" consists of halving block 
addresses, that is higher bit of block address lines must be set to low. 
I wonder if this strategy prevents paired page related issues. I think 
one should work at page addressing level instead, in order to use pages 
that are not paired only. Do you think it makes sense?

Andrea