Return to xfs_ag.h CVS log

Up to [Development] / xfs-linux-nodel

Lazy Superblock Counters

When we have a couple of hundred transactions on the fly at once,
they all typically modify the on disk superblock in some way.
create/unclink/mkdir/rmdir modify inode counts, allocation/freeing
modify free block counts.

When these counts are modified in a transaction, the must eventually
lock the superblock buffer and apply the mods.  The buffer then
remains locked until the transaction is committed into the incore
log buffer. The result of this is that with enough transactions on
the fly the incore superblock buffer becomes a bottleneck.

The result of contention on the incore superblock buffer is that
transaction rates fall - the more pressure that is put on the
superblock buffer, the slower things go.

The key to removing the contention is to not require the superblock
fields in question to be locked. We do that by not marking the
superblock dirty in the transaction. IOWs, we modify the incore
superblock but do not modify the cached superblock buffer. In short,
we do not log superblock modifications to critical fields in the
superblock on every transaction. In fact we only do it just before
we write the superblock to disk every sync period or just before
unmount.

This creates an interesting problem - if we don't log or write out
the fields in every transaction, then how do the values get
recovered after a crash? the answer is simple - we keep enough
duplicate, logged information in other structures that we can
reconstruct the correct count  after log recovery has been
performed.

It is the AGF and AGI structures that contain the duplicate
information; after recovery, we walk every AGI and AGF and sum their
individual counters to get the correct value, and we do a
transaction into the log to correct them. An optimisation of this is
that if we have a clean unmount record, we know the value in the
superblock is correct, so we can avoid the summation walk under
normal conditions and so mount/recovery times do not change under
normal operation.

One wrinkle that was discovered during development was that the
blocks used in the freespace btrees are never accounted for in the
AGF counters. This was once a valid optimisation to make; when the
filesystem is full, the free space btrees are empty and consume no
space. Hence when it matters, the "accounting" is correct.  But that
means the when we do the AGF summations, we would not have a correct
count and xfs_check would complain.  Hence a new counter was added
to track the number of blocks used by the free space btrees. This is
an *on-disk format change*.

As a result of this, lazy superblock counters are a mkfs option
and at the moment on linux there is no way to convert an old
filesystem. This is possible - xfs_db can be used to twiddle the
right bits and then xfs_repair will do the format conversion
for you. Similarly, you can convert backwards as well. At some point
we'll add functionality to xfs_admin to do the bit twiddling
easily....
Merge of xfs-linux-melb:xfs-kern:28652a by kenmcd.

  Changes to support lazy superblock counters.

/*
 * Copyright (c) 2000-2003,2005 Silicon Graphics, Inc.
 * All Rights Reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public License as
 * published by the Free Software Foundation.
 *
 * This program is distributed in the hope that it would be useful,
 * but WITHOUT ANY WARRANTY; without even the implied warranty of
 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 * GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
 * along with this program; if not, write the Free Software Foundation,
 * Inc.,  51 Franklin St, Fifth Floor, Boston, MA  02110-1301  USA
 */
#ifndef __XFS_AG_H__
#define	__XFS_AG_H__

/*
 * Allocation group header
 * This is divided into three structures, placed in sequential 512-byte
 * buffers after a copy of the superblock (also in a 512-byte buffer).
 */

struct xfs_buf;
struct xfs_mount;
struct xfs_trans;

#define	XFS_AGF_MAGIC	0x58414746	/* 'XAGF' */
#define	XFS_AGI_MAGIC	0x58414749	/* 'XAGI' */
#define	XFS_AGF_VERSION	1
#define	XFS_AGI_VERSION	1

#define	XFS_AGF_GOOD_VERSION(v)	((v) == XFS_AGF_VERSION)
#define	XFS_AGI_GOOD_VERSION(v)	((v) == XFS_AGI_VERSION)

/*
 * Btree number 0 is bno, 1 is cnt.  This value gives the size of the
 * arrays below.
 */
#define	XFS_BTNUM_AGF	((int)XFS_BTNUM_CNTi + 1)

/*
 * The second word of agf_levels in the first a.g. overlaps the EFS
 * superblock's magic number.  Since the magic numbers valid for EFS
 * are > 64k, our value cannot be confused for an EFS superblock's.
 */

typedef struct xfs_agf {
	/*
	 * Common allocation group header information
	 */
	__be32		agf_magicnum;	/* magic number == XFS_AGF_MAGIC */
	__be32		agf_versionnum;	/* header version == XFS_AGF_VERSION */
	__be32		agf_seqno;	/* sequence # starting from 0 */
	__be32		agf_length;	/* size in blocks of a.g. */
	/*
	 * Freespace information
	 */
	__be32		agf_roots[XFS_BTNUM_AGF];	/* root blocks */
	__be32		agf_spare0;	/* spare field */
	__be32		agf_levels[XFS_BTNUM_AGF];	/* btree levels */
	__be32		agf_spare1;	/* spare field */
	__be32		agf_flfirst;	/* first freelist block's index */
	__be32		agf_fllast;	/* last freelist block's index */
	__be32		agf_flcount;	/* count of blocks in freelist */
	__be32		agf_freeblks;	/* total free blocks */
	__be32		agf_longest;	/* longest free space */
	__be32		agf_btreeblks;	/* # of blocks held in AGF btrees */
} xfs_agf_t;

#define	XFS_AGF_MAGICNUM	0x00000001
#define	XFS_AGF_VERSIONNUM	0x00000002
#define	XFS_AGF_SEQNO		0x00000004
#define	XFS_AGF_LENGTH		0x00000008
#define	XFS_AGF_ROOTS		0x00000010
#define	XFS_AGF_LEVELS		0x00000020
#define	XFS_AGF_FLFIRST		0x00000040
#define	XFS_AGF_FLLAST		0x00000080
#define	XFS_AGF_FLCOUNT		0x00000100
#define	XFS_AGF_FREEBLKS	0x00000200
#define	XFS_AGF_LONGEST		0x00000400
#define	XFS_AGF_BTREEBLKS	0x00000800
#define	XFS_AGF_NUM_BITS	12
#define	XFS_AGF_ALL_BITS	((1 << XFS_AGF_NUM_BITS) - 1)

/* disk block (xfs_daddr_t) in the AG */
#define XFS_AGF_DADDR(mp)	((xfs_daddr_t)(1 << (mp)->m_sectbb_log))
#define	XFS_AGF_BLOCK(mp)	XFS_HDR_BLOCK(mp, XFS_AGF_DADDR(mp))
#define	XFS_BUF_TO_AGF(bp)	((xfs_agf_t *)XFS_BUF_PTR(bp))


/*
 * Size of the unlinked inode hash table in the agi.
 */
#define	XFS_AGI_UNLINKED_BUCKETS	64

typedef struct xfs_agi {
	/*
	 * Common allocation group header information
	 */
	__be32		agi_magicnum;	/* magic number == XFS_AGI_MAGIC */
	__be32		agi_versionnum;	/* header version == XFS_AGI_VERSION */
	__be32		agi_seqno;	/* sequence # starting from 0 */
	__be32		agi_length;	/* size in blocks of a.g. */
	/*
	 * Inode information
	 * Inodes are mapped by interpreting the inode number, so no
	 * mapping data is needed here.
	 */
	__be32		agi_count;	/* count of allocated inodes */
	__be32		agi_root;	/* root of inode btree */
	__be32		agi_level;	/* levels in inode btree */
	__be32		agi_freecount;	/* number of free inodes */
	__be32		agi_newino;	/* new inode just allocated */
	__be32		agi_dirino;	/* last directory inode chunk */
	/*
	 * Hash table of inodes which have been unlinked but are
	 * still being referenced.
	 */
	__be32		agi_unlinked[XFS_AGI_UNLINKED_BUCKETS];
} xfs_agi_t;

#define	XFS_AGI_MAGICNUM	0x00000001
#define	XFS_AGI_VERSIONNUM	0x00000002
#define	XFS_AGI_SEQNO		0x00000004
#define	XFS_AGI_LENGTH		0x00000008
#define	XFS_AGI_COUNT		0x00000010
#define	XFS_AGI_ROOT		0x00000020
#define	XFS_AGI_LEVEL		0x00000040
#define	XFS_AGI_FREECOUNT	0x00000080
#define	XFS_AGI_NEWINO		0x00000100
#define	XFS_AGI_DIRINO		0x00000200
#define	XFS_AGI_UNLINKED	0x00000400
#define	XFS_AGI_NUM_BITS	11
#define	XFS_AGI_ALL_BITS	((1 << XFS_AGI_NUM_BITS) - 1)

/* disk block (xfs_daddr_t) in the AG */
#define XFS_AGI_DADDR(mp)	((xfs_daddr_t)(2 << (mp)->m_sectbb_log))
#define	XFS_AGI_BLOCK(mp)	XFS_HDR_BLOCK(mp, XFS_AGI_DADDR(mp))
#define	XFS_BUF_TO_AGI(bp)	((xfs_agi_t *)XFS_BUF_PTR(bp))

/*
 * The third a.g. block contains the a.g. freelist, an array
 * of block pointers to blocks owned by the allocation btree code.
 */
#define XFS_AGFL_DADDR(mp)	((xfs_daddr_t)(3 << (mp)->m_sectbb_log))
#define	XFS_AGFL_BLOCK(mp)	XFS_HDR_BLOCK(mp, XFS_AGFL_DADDR(mp))
#define XFS_AGFL_SIZE(mp)	((mp)->m_sb.sb_sectsize / sizeof(xfs_agblock_t))
#define	XFS_BUF_TO_AGFL(bp)	((xfs_agfl_t *)XFS_BUF_PTR(bp))

typedef struct xfs_agfl {
	__be32		agfl_bno[1];	/* actually XFS_AGFL_SIZE(mp) */
} xfs_agfl_t;

/*
 * Busy block/extent entry.  Used in perag to mark blocks that have been freed
 * but whose transactions aren't committed to disk yet.
 */
typedef struct xfs_perag_busy {
	xfs_agblock_t	busy_start;
	xfs_extlen_t	busy_length;
	struct xfs_trans *busy_tp;	/* transaction that did the free */
} xfs_perag_busy_t;

/*
 * Per-ag incore structure, copies of information in agf and agi,
 * to improve the performance of allocation group selection.
 *
 * pick sizes which fit in allocation buckets well
 */
#if (BITS_PER_LONG == 32)
#define XFS_PAGB_NUM_SLOTS	84
#elif (BITS_PER_LONG == 64)
#define XFS_PAGB_NUM_SLOTS	128
#endif

typedef struct xfs_perag
{
	char		pagf_init;	/* this agf's entry is initialized */
	char		pagi_init;	/* this agi's entry is initialized */
	char		pagf_metadata;	/* the agf is preferred to be metadata */
	char		pagi_inodeok;	/* The agi is ok for inodes */
	__uint8_t	pagf_levels[XFS_BTNUM_AGF];
					/* # of levels in bno & cnt btree */
	__uint32_t	pagf_flcount;	/* count of blocks in freelist */
	xfs_extlen_t	pagf_freeblks;	/* total free blocks */
	xfs_extlen_t	pagf_longest;	/* longest free space */
	__uint32_t	pagf_btreeblks;	/* # of blocks held in AGF btrees */
	xfs_agino_t	pagi_freecount;	/* number of free inodes */
	xfs_agino_t	pagi_count;	/* number of allocated inodes */
	int		pagb_count;	/* pagb slots in use */
#ifdef __KERNEL__
	lock_t		pagb_lock;	/* lock for pagb_list */
#endif
	xfs_perag_busy_t *pagb_list;	/* unstable blocks */
} xfs_perag_t;

#define	XFS_AG_MAXLEVELS(mp)		((mp)->m_ag_maxlevels)
#define	XFS_MIN_FREELIST_RAW(bl,cl,mp)	\
	(MIN(bl + 1, XFS_AG_MAXLEVELS(mp)) + MIN(cl + 1, XFS_AG_MAXLEVELS(mp)))
#define	XFS_MIN_FREELIST(a,mp)		\
	(XFS_MIN_FREELIST_RAW(		\
		be32_to_cpu((a)->agf_levels[XFS_BTNUM_BNOi]), \
		be32_to_cpu((a)->agf_levels[XFS_BTNUM_CNTi]), mp))
#define	XFS_MIN_FREELIST_PAG(pag,mp)	\
	(XFS_MIN_FREELIST_RAW(		\
		(uint_t)(pag)->pagf_levels[XFS_BTNUM_BNOi], \
		(uint_t)(pag)->pagf_levels[XFS_BTNUM_CNTi], mp))

#define XFS_AGB_TO_FSB(mp,agno,agbno)	\
	(((xfs_fsblock_t)(agno) << (mp)->m_sb.sb_agblklog) | (agbno))
#define	XFS_FSB_TO_AGNO(mp,fsbno)	\
	((xfs_agnumber_t)((fsbno) >> (mp)->m_sb.sb_agblklog))
#define	XFS_FSB_TO_AGBNO(mp,fsbno)	\
	((xfs_agblock_t)((fsbno) & XFS_MASK32LO((mp)->m_sb.sb_agblklog)))
#define	XFS_AGB_TO_DADDR(mp,agno,agbno)	\
	((xfs_daddr_t)XFS_FSB_TO_BB(mp, \
		(xfs_fsblock_t)(agno) * (mp)->m_sb.sb_agblocks + (agbno)))
#define	XFS_AG_DADDR(mp,agno,d)		(XFS_AGB_TO_DADDR(mp, agno, 0) + (d))

/*
 * For checking for bad ranges of xfs_daddr_t's, covering multiple
 * allocation groups or a single xfs_daddr_t that's a superblock copy.
 */
#define	XFS_AG_CHECK_DADDR(mp,d,len)	\
	((len) == 1 ? \
	    ASSERT((d) == XFS_SB_DADDR || \
		   XFS_DADDR_TO_AGBNO(mp, d) != XFS_SB_DADDR) : \
	    ASSERT(XFS_DADDR_TO_AGNO(mp, d) == \
		   XFS_DADDR_TO_AGNO(mp, (d) + (len) - 1)))

#endif	/* __XFS_AG_H__ */