2008-06-01 20:38:54

by Miklos Szeredi

Subject: Re: [patch 3/8] Make d_path() consistent across mount operations

> > The path that __d_path() computes can become slightly inconsistent when it
> > races with mount operations: it grabs the vfsmount_lock when traversing mount
> > points but immediately drops it again, only to re-grab it when it reaches the
> > next mount point. The result is that the filename computed is not always
> > consisent, and the file may never have had that name. (This is unlikely, but
> > still possible.)
> >
> > Fix this by grabbing the vfsmount_lock for the whole duration of
> > __d_path().
>
> Looks good, and lock holding times shouldn't be a problem either.

Thanks for the review of this batch.

Can you please in the future either explicitly ACK or NACK? Because I
really wouldn't want any more of this "Looks good, but bla bla bla"
and then NACKing the patch when I send it off to Andrew.

Miklos

2008-06-02 05:56:03

by Christoph Hellwig

Subject: Re: [patch 3/8] Make d_path() consistent across mount operations

On Sun, Jun 01, 2008 at 10:38:35PM +0200, Miklos Szeredi wrote:
> > > The path that __d_path() computes can become slightly inconsistent when it
> > > races with mount operations: it grabs the vfsmount_lock when traversing mount
> > > points but immediately drops it again, only to re-grab it when it reaches the
> > > next mount point. The result is that the filename computed is not always
> > > consisent, and the file may never have had that name. (This is unlikely, but
> > > still possible.)
> > >
> > > Fix this by grabbing the vfsmount_lock for the whole duration of
> > > __d_path().
> >
> > Looks good, and lock holding times shouldn't be a problem either.
>
> Thanks for the review of this batch.
>
> Can you please in the future either explicitly ACK or NACK? Because I
> really wouldn't want any more of this "Looks good, but bla bla bla"
> and then NACKing the patch when I send it off to Andrew.

ACK for patches 1-3.