[ECOS] JFFS2 Garbage Collection Error? (NO! rbtree.c bug)

[David Woodhouse]

On Mon, 2003-07-14 at 19:26, Scott Wilkinson wrote:

> Actually, there is one other thing that I had to do to get my tests to
> run.  Perhaps David can comment...
>  I had to comment out a check in the
> JFFS2 read.c file.  I am not sure what it was doing, but I can't run
> with it in.
> 
> Take a look at the code below.  Hopefully my mail client will not munge
> it too hard.  Look for my "#if 0"

Your fix is correct. It's an over-zealous check for overlaps in the
extent tree for the file, which is designed to trigger any time you read
from anywhere other than the beginning of an extent (or 'fragment'),
except if you're at the beginning of a page. This is because we know the
Linux code, operating via the page cache, would _always_ start reading
at the beginning of a page. 

That assumption isn't valid any more since eCos reads from arbitrary
offsets, and the check is for a case which never happened and would make
itself known elsewhere too anyway -- I've just removed it from CVS.

Thanks.

Would someone care to run the 'mkepk.sh' script from current CVS code
and commit the result to the eCos tree?

-- 
dwmw2


-- 
Before posting, please read the FAQ: http://sources.redhat.com/fom/ecos
and search the list archive: http://sources.redhat.com/ml/ecos-discuss