RE: [PATCH] Obsolete nodes that are unlinked when possible

["Joakim Tjernlund" <joakim.tjernlund@transmode.se>]

> -----Original Message-----
> From: David Woodhouse [mailto:dwmw2@infradead.org] 
> Sent: den 20 april 2007 23:14
> To: Joakim Tjernlund
> Cc: linux-mtd@lists.infradead.org
> Subject: RE: [PATCH] Obsolete nodes that are unlinked when possible
> 
> On Thu, 2007-04-19 at 21:34 +0200, Joakim Tjernlund wrote:
> > Ping. 
> 
> Sorry for taking so long to look at this. Did you test this without
> SUMMARY support?

hmm, not sure. I think so, but not much. The idea was that to get
some testing, you need to enable its use in MTD git. 
 
> 
> I agree that we ought to make the summary code _not_ imply
> !jffs2_can_mark_obsolete(). There's a few things I want to 
> change about
> the summary code -- the nodes are currently about twice as 
> large as they
> need to be, too. I'll look at that soon.

Rigth, but more important is that the GC procedure needs to GC
more aggressively, otherwise you will end up with a very fragmented
FS in some cases.

ATM I am doing board bringup and won't be able to spend any time
on JFFS2.

  Jocke