Re: [prepatch] address_space-based writeback

[Jan Harkes]

On Fri, Apr 12, 2002 at 08:57:17AM +0100, Anton Altaparmakov wrote:
> Yet, this really begs the question of defining the concept of a file. I am 
> quite happy with files being the io entity in ntfs. It is just that each 
> file in ntfs can contain loads of different data holding attributes which 
> are all worth placing in address spaces. Granted, a dummy inode could be 
> setup for each of those which just means a lot of wasted ram but ntfs is 
> not that important so I have to take the penalty there. But if I also need 
> unique inode numbers in those dummy inodes then the overhead is becoming 
> very, very high...

You could have all additional IO streams use the same inode number and
use iget4. Several inodes can have the same i_ino and the additional
argument would be a stream identifier that selects the correct 'IO
identity'.

Jan