> Alexander Viro <viro@math.psu.edu> wrote:
> > yes, FAT is in sore need of pointed stick, mallet and garlic; no, even its
> > authors can't kill the bloody thing and they would *like* to do that.
>
> Consider a mount-time option supporting a file which maps each file
> name to an inode number. You'd have to peg the entire file system
> directory tree into memory on mount (and post big warnings about
> the failure modes), but maybe it would be good.
>
> Not saying that you should implement it, just critique the
> idea.
Feel free to implement it. IOW, it's going to be choke-full of races. I
don't envy anybody who will try to do it. Modifications of said file upon
rename()... <shudder> Besides, you'll have to do search in that file.
Definitely not my idea of fun. There are simpler ways to implement a
decent filesystem at *that* cost - after all, since you are going to keep
a lot of metadata in-core you may as well try to write vfat<->ext2
convertor ;-)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/