Re: [PATCH 09/10] mm: expose BDI statistics in sysfs.

[Peter Zijlstra <a.p.zijlstra@chello.nl>]

On Sat, 2007-04-21 at 02:55 -0700, Andrew Morton wrote:
> On Fri, 20 Apr 2007 17:52:03 +0200 Peter Zijlstra <a.p.zijlstra@chello.nl> wrote:
> 
> > Expose the per BDI stats in /sys/block/<dev>/queue/*
> > 
> > Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
> > ---
> >  block/ll_rw_blk.c |   32 ++++++++++++++++++++++++++++++++
> >  1 file changed, 32 insertions(+)
> > 
> > Index: linux-2.6-mm/block/ll_rw_blk.c
> > ===================================================================
> > --- linux-2.6-mm.orig/block/ll_rw_blk.c
> > +++ linux-2.6-mm/block/ll_rw_blk.c
> > @@ -3976,6 +3976,15 @@ static ssize_t queue_max_hw_sectors_show
> >  	return queue_var_show(max_hw_sectors_kb, (page));
> >  }
> >  
> > +static ssize_t queue_nr_reclaimable_show(struct request_queue *q, char *page)
> > +{
> > +	return sprintf(page, "%lld\n", bdi_stat(&q->backing_dev_info, BDI_RECLAIMABLE));
> > +}
> 
> We try to present memory statistics to userspace in bytes or kbytes rather
> than number-of-pages.  Because page-size varies between architectures and
> between .configs.  Displaying number-of-pages is just inviting people to write
> it-broke-when-i-moved-it-to-ia64 applications.
> 
> Plus kbytes is a bit more user-friendly, particularly when the user will
> want to compare these numbers to /proc/meminfo, for example.
> 
> Using %llu might be more appropriate than %lld.


Right, the biggest problem I actually have with his piece of code is
that is does not represent all BDIs in the system. For example, the BDI
of NFS mounts is not accessible.

Will fix the other points.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>