Re: compat: autofs v5 packet size ambiguity - update

[Ian Kent]

On Wed, 2012-02-22 at 20:15 +0800, Ian Kent wrote:
> On Wed, 2012-02-22 at 17:32 +0800, Ian Kent wrote:
> > On Wed, 2012-02-22 at 13:57 +0800, Ian Kent wrote:
> > > On Wed, 2012-02-22 at 13:53 +0800, Ian Kent wrote:
> > > > On Wed, 2012-02-22 at 13:43 +0800, Ian Kent wrote:
> > > > > On Tue, 2012-02-21 at 20:56 -0800, Linus Torvalds wrote:
> > > > 
> > > > Ahh ... forgot to set the file_operations structure member .. oops
> > > > 
> > > > >  
> > > > > +static int autofs4_root_dir_open(struct inode *inode, struct file *file)
> > > > > +{
> > > > > +	struct autofs_sb_info *sbi= autofs4_sbi(file->f_path.dentry->d_sb);
> > > > > +	if (sbi->compat_daemon < 0)
> > > > > +		sbi->compat_daemon = is_compat_task();
> > > > > +	return dcache_dir_open(inode, file);
> > > > > +}
> > > > > +
> > > > 
> > > 
> > > Lets try that again.
> > 
> > Oh, DOH ... 
> > 
> > The process mounting the autofs mount is passing the pipe it will use
> > for communication via a mount option so we can do this right in
> > fill_super.
> > 
> > I'm out for a while, I'll update the patch when I'm back.

This is *still* not right.

The daemon could be SIGKILLed and then re-connect to the existing mounts
and set a new pipe file handle. But that new task may or may not be a
compat task.

Let me fix that too.



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