Luc Van Oostenryck (privately) wrote:

>> Here's question related to my JFFS2 problems (posted on the
>> MTD-list).
>> 
>> The situation: The mount times of my 32 megs JFFS2 device suddenly 
>> increased from seconds to (roughly) 9 minutes, after having written
>>  some files to the device. The explanation of this is, according to
>> David Woodhouse's best guess, that the garbage collector uses all
>> this time "for building up the node tree for every inode after
>> mounting". The problem stems (I've been told) from the fact that I
>> have been performing "big-file-gymnastics" (11 megs uncompressed -
>> ~3 megs compressed) on the device. Possibly along with some
>> small-file actions inbetween (?)...
>> 
>> Now my question: Can YAFFS (1&2?) be provoked into showing similar 
>> "unfortunate" behaviour, or is it handled in another way?

> I don't know very well JFFS2, but no, I have never seen this with
> YAFFS. The scanning time (=~ mount time) of YAFFS is never slow like
> this: the maximum mount time is linearly bounded to the size and the
> speed of the device.
And hopefully with a small gradient...

> But the speed can can be difficult to really compare with JFFS2, 
> since JFFS(2) run generally on NOR flash while YAFFS run generally
> run on NAND flash for which it was designed). Also, YAFFS don't need
> to run it's garbage collector at mount time and do not have
> compressed files.
Actually, JFFS2 does not neither - the initial building/scanning of the 
node-tree is just performed by the same process...
But I think the trend at the moment it is to tweak JFFS2 against NAND 
usage (http://www.inf.u-szeged.hu/jffs2/mount.php)...

However, I see the point in using a filesystem originally tailored for 
NAND...

I just need making sure I can have some ECC enabled - either in MTD or 
YAFFS...

// Martin