JFFS3 memory consumption

[Thomas Gleixner]

On Thu, 2005-01-27 at 22:35 +0000, David Woodhouse wrote:
> On Thu, 2005-01-27 at 21:23 +0100, Thomas Gleixner wrote:
> > Keep it write-through !
> > 
> > Write caching is contrary to powerfail robustness.
> 
> No. O_SYNC, fsync() and sync() are there for a reason, and even on NAND
> today you have to use them correctly according to POSIX if you want sane
> results. We're not pandering to stupidity.

Sure. I'm aware of O_SYNC, fsync() and sync().

Please reread http://sources.redhat.com/jffs2/jffs2.pdf

tglx