Re: [PATCH v2] procfs: fix numbering in /proc/locks

[Andrew Morton <akpm@linux-foundation.org>]

On Tue, 05 Oct 2010 14:14:11 +0200
Jerome Marchand <jmarchan@redhat.com> wrote:

> On 09/30/2010 02:38 PM, Jerome Marchand wrote:
> > 
> > The lock number in /proc/locks (first field) is implemented by a counter
> > (private field of struct seq_file) which is incremented at each call of
> > locks_show() and reset to 1 in locks_start() whatever the offset is. It
> > should be reset according to the actual position in the list.
> > 
> > Moreover, locks_show() can be called twice to print a single line thus
> > skipping a number. The counter should be incremented in locks_next().
> > 
> > Signed-off-by: Jerome Marchand <jmarchan@redhat.com>
> > ---
> >  locks.c |    4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/fs/locks.c b/fs/locks.c
> > index ab24d49..49d7343 100644
> > --- a/fs/locks.c
> > +++ b/fs/locks.c
> > @@ -2166,19 +2166,19 @@ static int locks_show(struct seq_file *f, void *v)
> >  	list_for_each_entry(bfl, &fl->fl_block, fl_block)
> >  		lock_get_status(f, bfl, (long)f->private, " ->");
> >  
> > -	f->private++;
> >  	return 0;
> >  }
> >  
> >  static void *locks_start(struct seq_file *f, loff_t *pos)
> >  {
> >  	lock_kernel();
> > -	f->private = (void *)1;
> > +	f->private = (void *) (*pos + 1);
> 
> That cast trigger a warning on some arch:
> "warning: cast to pointer from integer of different size"
> 
> There is no real risk here. At worst /proc/locks will show wrong number
> if there is more than 2^32 locks, but should I mute the warning it with
> something like:
>         f->private = (void *) (size_t) (*pos + 1);
> ?

Putting a loff_t into a void* is a pretty alarming thing to do.  If
we're really going to do that then use a (long) cast and put a very
good comment at the code site explaining why the bug doesn't matter, so
people aren't misled.

But really, why sweat it?  kmalloc the eight bytes, make ->private
point at that and we never have to think about it again.  Bonus points
for doing this without any typecasts ;)