lkml.org 
[lkml]   [1998]   [Jan]   [5]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: dcache strikes back
> this thread about vfat/dcache problems made me think about
> the affs as well, and so I just tried it. As I was already
> expecting, it failed. :-(
>
> /root# echo blafasel >/amiga/test/file
> /root# ls -l /amiga/test/FILE
> -rw-r--r-- 1 root root 9 Jan 5 17:55 /amiga/test/FILE
> /root# rm /amiga/test/file
> /root# ls -l /amiga/test/FILE
> -rw-r--r-- 0 root root 9 Jan 5 17:55 /amiga/test/FILE
>
> Obviously, vfat does _this_ right, but by looking at the code,
> I just don't see how. Isn't that just the same problem as the
> longname/shortname one, multiple dentries for a single inode?

The situation with case insensitivity is much easier to handle than the
short-name aliases required for vfat. Basically you just need a suitable
dentry_ops with d_hash and d_compare. This will ensure that only one
dentry for a given file is created.

There are examples of the d_op routines in ncpfs and smbfs as well as
the fat fs. If affs has only a total length restriction on the name
structure, the model in ncpfs or smbfs would probably be closer to what
you need.

Regards,
Bill

\
 
 \ /
  Last update: 2005-03-22 13:41    [W:0.043 / U:0.200 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site