Re: [PATCH v4 2/3] xfs: introduce in-core global counter of allocbt blocks

[Brian Foster <bfoster@redhat.com>]

On Tue, Apr 27, 2021 at 03:58:16PM +0530, Chandan Babu R wrote:
> On 23 Apr 2021 at 18:40, Brian Foster wrote:
> > Introduce an in-core counter to track the sum of all allocbt blocks
> > used by the filesystem. This value is currently tracked per-ag via
> > the ->agf_btreeblks field in the AGF, which also happens to include
> > rmapbt blocks. A global, in-core count of allocbt blocks is required
> > to identify the subset of global ->m_fdblocks that consists of
> > unavailable blocks currently used for allocation btrees. To support
> > this calculation at block reservation time, construct a similar
> > global counter for allocbt blocks, populate it on first read of each
> > AGF and update it as allocbt blocks are used and released.
> >
> > Signed-off-by: Brian Foster <bfoster@redhat.com>
> > ---
> >  fs/xfs/libxfs/xfs_alloc.c       | 12 ++++++++++++
> >  fs/xfs/libxfs/xfs_alloc_btree.c |  2 ++
> >  fs/xfs/xfs_mount.h              |  6 ++++++
> >  3 files changed, 20 insertions(+)
> >
> > diff --git a/fs/xfs/libxfs/xfs_alloc.c b/fs/xfs/libxfs/xfs_alloc.c
> > index aaa19101bb2a..144e2d68245c 100644
> > --- a/fs/xfs/libxfs/xfs_alloc.c
> > +++ b/fs/xfs/libxfs/xfs_alloc.c
> > @@ -3036,6 +3036,7 @@ xfs_alloc_read_agf(
> >  	struct xfs_agf		*agf;		/* ag freelist header */
> >  	struct xfs_perag	*pag;		/* per allocation group data */
> >  	int			error;
> > +	uint32_t		allocbt_blks;
> >
> >  	trace_xfs_alloc_read_agf(mp, agno);
> >
> > @@ -3066,6 +3067,17 @@ xfs_alloc_read_agf(
> >  		pag->pagf_refcount_level = be32_to_cpu(agf->agf_refcount_level);
> >  		pag->pagf_init = 1;
> >  		pag->pagf_agflreset = xfs_agfl_needs_reset(mp, agf);
> > +
> > +		/*
> > +		 * Update the global in-core allocbt block counter. Filter
> > +		 * rmapbt blocks from the on-disk counter because those are
> > +		 * managed by perag reservation.
> > +		 */
> > +		if (pag->pagf_btreeblks > be32_to_cpu(agf->agf_rmap_blocks)) {
> 
> pag->pagf_btreeblks gets incremented everytime a block is allocated to refill
> AGFL (via xfs_alloc_get_freelist()). Apart from the allobt trees, blocks for
> Rmap btree also get allocated from AGFL. Hence pag->pagf_btreeblks must be
> larger than agf->agf_rmap_blocks.
> 

This function is actually to consume a block from the AGFL (as opposed
to refill it).

> Can you please describe the scenario in which pag->pagf_btreeblks has a value
> that is <= agf->agf_rmap_blocks?
> 

Ah, this was just an initialization quirk. I originally had an assert
here and based the logic on the assumption that pagf_btreeblks >=
agf_rmap_blocks, but alas:

# mkfs.xfs -f -mrmapbt <dev>
...
# xfs_db -c "agf 0" -c "p rmapblocks" -c "p btreeblks" <dev>
rmapblocks = 1
btreeblks = 0
#

Brian

> --
> chandan
>