Re: [PATCH] ramfs: pretend dirent sizes

[Jan Blunck <j.blunck@tu-harburg.de>]

Chris Wedgwood wrote:
> 
>>To my understanding, you can lseek to any "proper" offset inside a
>>directory.  Proper means that the offset marks the beginning of a
>>new dirent (or end of file) in the interpretation of the filesystem.
> 
> 
> But you can never tell where these are in general.
> 
> 

I don't want to tell where these are in general, I need an easy way to 
seek to the m'th directory + offset position without reading every 
single dirent. With i_sizes != 0 it is straight forward to use "the sum 
of the m directory's i_sizes + offset" as the f_pos to seek to. For this 
purpose it is not necessary to have a "honest" i_size as long as the 
i_size is bigger than the offset of the last dirent in the directory.

> 
> I don't see why or how this can be true in general (it might be, but I
> don't see how myself).  If we are half way through scanning a
> directory and people start messing with it we could end up somewhere
> bogus (in which case f_op->readdir I guess is expected to try and do
> something sane here?)
> 

E.g.: ext2_validate_entry() is called if filp->f_version != 
inode->i_version.

> 
>>Reopening the same directory may result in a formerly proper offset
>>isn't anymore.
> 

Which is a user problem again. Might be that you are opening a different 
directory with the same name ... or even a regular file!

Jan