inode checkpoints

[David Woodhouse]

On Mon, 2004-10-04 at 07:36 -0500, Josh Boyer wrote:
> Yes and no.  It makes sense for any flash device that you can't write to
> the same page multiple times without an erase in between.  NAND is
> certainly the largest group of such devices.  There are others that have
> this behavior as well.  See cfi_cmdset_0020.c.
> 
> Although, JFFS2 doesn't support those chips right now, so I guess I am
> just rambling :).

There's more chips like that on the way too. It'll be 

	if (!jffs2_can_mark_obsolete(c))

and it'll probably also be in the ifdef, although we may actually change
the name of that configuration option.

I agree that it shouldn't be left to the GC thread alone -- there isn't
one on eCos and it's optional on Linux too. We can write out snapshots
in the normal course of garbage collection though, when we have
obsoleted a node belonging to the inode in question.



-- 
dwmw2