Re: [patch 13/13] lseek speedup

[Anton Altaparmakov]

At 17:49 17/07/02, Andrew Morton wrote:
>Anton Altaparmakov wrote:
> >
> > >  Now, why are we taking i_sem for lseek/readdir
> > >exclusion and not a per-file lock?
> >
> > Because it also excludes against directory modifications, etc. Just imagine
> > what "rm somefile" or "mv somefile otherfile" or "touch newfile" would do
> > to the directory contents and what a concurrent readdir() would do... A
> > very loud *BANG* is the only thing that springs to mind...
>
>That's different.  i_size, contents of things, yes - i_sem for
>those.
>
>But protection of struct file should not be via any per-inode thing.

Oh, I see. But that would mean adding yet another lock to an existing 
locking scheme? So both i_sem and the "per file lock" would nead to be held 
over readdir(). That's doable but it would have to be a semaphore based 
lock due to readdir()...

Perhaps the f_lock from your patch, changed to a semaphore, and acquiring 
it in generic_file_llseek&friends, vfs_readdir() (and any other place where 
f_pos needs protecting) would be able to do the trick. Would that solve the 
lock contention problems you were seing? It would at least off-load i_sem a 
bit... but it would only replace one semaphore for another...

> > btw. the directory modification locking rules are written up in
> > Documentation/filesystems/directory-locking by our very own VFS maintainer
> > Al Viro himself... (-;
>
>Doesn't cover lseek...

Hm, true. I hadn't read it for quite a while...

         Anton