inode clashes (linux 2.2.x, arla 0.2x)

[Magnus Ahltorp]

> I just discovered that using fully qualified file names tend to give
> every file in a directory the same inode number:

The inum is used in both readdir and stat. When doing stat, the inum
will be the pointer of the struct xnode. When doing readdir, there are
two possiblities:

* The file is in the xfs cache. The inum presented will be the same as
  when doing stat.
* The file is not in the xfs cache. The inum presented will be the
  inum calculated by arlad.

> Is it possible to use the same algorithm as the other major afs client
> vendor does? It seem to produce unique inodes even across platforms.

The problem with that algorithm is that (machine, inum)->file is not
guaranteed to be a function, which is a requirement for Linux.

/Magnus
map at stacken.kth.se