[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Expected behavior of JFFS?



David Woodhouse wrote:

> If you just enable the counting of no_jffs_file unconditionally, then you
> get that information for nothing. We've already done the same thing for
> no_jffs_node.

Right. It was quick and dirty to see if it was working. Laziness on my part... I'll
fix and deposit (without messing the indentation this time).

>
> Perhaps you could keep a record for each file of whether it's stored
> optimally or not. And a global count of the number of files which are in
> this suboptimal state. If that count exceeds a certain percentage of the
> number of files, then trigger GC?

How about directing GC, or more precisly jffs_rewrite_data(), to a file that needs
it badly? In other words, go on a per file basis, and if this file exceed a
node-to-size ratio, we "jffs_rewrite_data()" it?

Again, this thing is useful when the amount of node in use is extreme. We could
just increase the base limit (25) to 100, and GC would be less triggered, and Mark
would be happy too.

--
Martin Gadbois
S/W designer
Colubris Networks (http://www.colubris.com)




To unsubscribe from this list: send the line "unsubscribe jffs-dev" in
the body of a message to majordomo@xxxxxxx.com