Re: [PATCH 2/3] xfs: allow scrubbers to pause background reclaim

[Dave Chinner <david@fromorbit.com>]

On Wed, Apr 17, 2019 at 03:29:53PM -0700, Darrick J. Wong wrote:
> On Thu, Apr 18, 2019 at 07:52:28AM +1000, Dave Chinner wrote:
> > On Tue, Apr 16, 2019 at 06:40:12PM -0700, Darrick J. Wong wrote:
> > > From: Darrick J. Wong <darrick.wong@oracle.com>
> > > 
> > > The forthcoming summary counter patch races with regular filesystem
> > > activity to compute rough expected values for the counters.  This design
> > > was chosen to avoid having to freeze the entire filesystem to check the
> > > counters, but while that's running we'd prefer to minimize background
> > > reclamation activity to reduce the perturbations to the incore free
> > > block count.  Therefore, provide a way for scrubbers to disable
> > > background posteof and cowblock reclamation.
> > > 
> > > Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
> > > ---
> > >  fs/xfs/scrub/common.c |   18 ++++++++++++++++++
> > >  fs/xfs/scrub/common.h |    2 ++
> > >  fs/xfs/scrub/scrub.c  |    2 ++
> > >  fs/xfs/scrub/scrub.h  |    1 +
> > >  4 files changed, 23 insertions(+)
> > > 
> > > 
> > > diff --git a/fs/xfs/scrub/common.c b/fs/xfs/scrub/common.c
> > > index 7076d5c98151..a406a22a734f 100644
> > > --- a/fs/xfs/scrub/common.c
> > > +++ b/fs/xfs/scrub/common.c
> > > @@ -894,3 +894,21 @@ xchk_ilock_inverted(
> > >  	}
> > >  	return -EDEADLOCK;
> > >  }
> > > +
> > > +/* Pause background reclamation and inactivation. */
> > > +void
> > > +xchk_disable_reclaim(
> > > +	struct xfs_scrub	*sc)
> > > +{
> > > +	sc->flags |= XCHK_RECLAIM_DISABLED;
> > > +	xfs_icache_disable_reclaim(sc->mp);
> > > +}
> > 
> > Hmmm. Poorly named function.  "reclaim" in the context of
> > xfs_icache.c means inode cache reclaim...
> > 
> > We're not actually disabling inode reclaim here, we
> > are disabling background block reaping.
> > 
> > Can we change this all to be named more appropriately (including the
> > icache.c functions? Then it is all fine because I don't look at
> > it and think "that'll break under memory pressure"....
> 
> Ah, reaping!  That's the word I was looking for. :(
> 
> xfs_icache_disable_speculative_reaping
> xchk_disable_speculative_reaping

disable_block_reaping?

but, really, purple or pink at this point...

Cheers,

Dave.
-- 
Dave Chinner
david@fromorbit.com