Eraseblocks torture: OneNAND results

[Ricard Wanderlof]

On Fri, 8 Dec 2006, Artem Bityutskiy wrote:

> 1. Although the documented eraseblock lifetime is about 100000, we 
> started getting errors after about 6 million erase cycles. May be 
> because our test is not torturous enough.

Isn't 100000 a figure quoted by the flash manufacturer to be the _minimum_ 
number of erase cycles, a specification which all devices must meet, even 
at extreme operating conditions (e.g. high temperature and extreme supply 
voltages)? That would imply that during more ordinary conditions, a random 
sample of flash chip would very likely be erasable many more times.

Another thing: Why implement it as a kernel module? Speed I would assume?

/Ricard