JFFS3 & performance

[Josh Boyer]

On Wed, 2005-01-12 at 16:55, Jared Hulbert wrote:
> 
> I respectfully disagree.  I don't think checksums are needed to
> protect you from NOR read errors *unless* the checksums are the only
> thing protecting the filesystem from bad things like crashes, power
> failures, and bugs.

Guess we'll have to agree to disagree then :).  All I know is that I
want to be damn sure that the data I'm returning isn't totally screwed. 
Call me paranoid.  A checksum is the only way I know of doing that.

David summed it up pretty good.  Crap happens in a multitude of ways. 
Even a bug in the write method of the filesystem could cause invalid
data to be written.  A checksum can at least help point stuff like that
out.

Poo is icky.  I'd rather try to go around it than jump in the middle of
it. :)