Re: [RFC] atomic create+open

[Miklos Szeredi <miklos@szeredi.hu>]

> > You can replace the inode in ->create_open() if you want to.
> 
> Thinking a bit more clearly after a cup of coffee. This statement isn't
> even true.
> 
> Your pseudo-code offers no guarantees that you are the sole user of the
> dentry once you get to create_open().

You are right.  I meant, replace the dentry.

> > Or let the VFS redo the lookup (as if d_revalidate() returned 0).
> 
> Which may return yet another result for the dentry and another race.
> There is no guarantee that you will ever make progress if someone is
> doing something like.
> 
> while true
> do
>   echo "1" > foo
>   echo "2" > foo
> done
> 
> on the server.

Not good example. This won't change the file, only the contents.
Something with rename would be better.

We are still pitting two different races against each other.  I can't
see such a big difference in ugliness...

Miklos



-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/