Re: [PATCH 6/8] autofs - use path_is_mountpoint() to fix unreliable d_mountpoint() checks

[Ian Kent]

On Thu, 2016-10-27 at 03:17 +0100, Al Viro wrote:
> On Tue, Oct 11, 2016 at 01:34:18PM +0800, Ian Kent wrote:
> > 
> > +	path = file->f_path;
> > +
> > Â	/*
> > Â	Â* An empty directory in an autofs file system is always a
> > Â	Â* mount point. The daemon must have failed to mount this
> > @@ -123,7 +126,7 @@ static int autofs4_dir_open(struct inode *inode, struct
> > file *file)
> > Â	Â* it.
> > Â	Â*/
> > Â	spin_lock(&sbi->lookup_lock);
> > -	if (!d_mountpoint(dentry) && simple_empty(dentry)) {
> > +	if (!path_is_mountpoint(&path) && simple_empty(dentry)) {
> Why not &file->f_path, provided that you constify that thing properly?

Yep, my bad, as pointed out by Eric.

Patches to fix that and constify a bunch of things will follow.

> 
> > 
> > +		if (rcu_walk) {
> > +			if (!path_is_mountpoint_rcu(path))
> > +				return -EISDIR;
> > +		} else {
> > +			if (!path_is_mountpoint(path))
> > +				return -EISDIR;
> IDGI.ÂÂWhat's the point of _having_ the _rcu() variant, anyway?ÂÂHere you
> are probably paying more in terms of i-cache footprint/branch prediction
> than you win on not doing that rcu_read_lock()/rcu_read_unlock()...
> 
> _rcu variants make sense when non-RCU case does something you can't do
> under RCU; here your path_is_mountpoint() is pretty close to being
> rcu_read_lock()+path_is_mountpoint_rcu()+rcu_read_unlock() anyway...

Again, my bad, I'll merge these two and post along with the follow up patches
above.

Thanks Al,
Ian