Re: [PATCH] VFS: allow ->d_manage() to declare -EISDIR in rcu_walk mode.

From: NeilBrown
Date: Fri Aug 01 2014 - 04:48:38 EST


On Fri, 1 Aug 2014 11:33:18 +0300 Dan Carpenter <dan.carpenter@xxxxxxxxxx>
wrote:

> On Wed, Jul 30, 2014 at 04:08:33PM +1000, NeilBrown wrote:
> > @@ -1110,11 +1110,16 @@ static bool __follow_mount_rcu(struct nameidata *nd, struct path *path,
> > * Don't forget we might have a non-mountpoint managed dentry
> > * that wants to block transit.
> > */
> > - if (unlikely(managed_dentry_might_block(path->dentry)))
> > + switch (managed_dentry_rcu(path->dentry)) {
> > + case -ECHILD:
> > + default:
> > return false;
> > + case -EISDIR:
> > + return true;
> > + }
> >
>
> Smatch says that any lines after that switch statement are unreachable.
>
> Is the "default" intended?

Ugh. It was an after thought.

The defined error codes are -ECHILD and -EISDIR.
But current code actually treats any other negative number as -ECHILD.
So I stuck in the default .... which of course was wrong.
I need to add
case 0:
break;

I'll send that out on Monday.

Thanks heaps.

NeilBrown


>
> > if (!d_mountpoint(path->dentry))
> > - return true;
> > + return !(path->dentry->d_flags & DCACHE_NEED_AUTOMOUNT);
> >
> > mounted = __lookup_mnt(path->mnt, path->dentry);
> > if (!mounted)
>
> regards,
> dan carpenter

Attachment: signature.asc
Description: PGP signature