92821e2ba4
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, they 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.... SGI-PV: 964999 SGI-Modid: xfs-linux-melb:xfs-kern:28652a Signed-off-by: David Chinner <dgc@sgi.com> Signed-off-by: Christoph Hellwig <hch@infradead.org> Signed-off-by: Tim Shimmin <tes@sgi.com>
165 lines
5.4 KiB
C
165 lines
5.4 KiB
C
/*
|
|
* Copyright (c) 2000,2005 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_IALLOC_H__
|
|
#define __XFS_IALLOC_H__
|
|
|
|
struct xfs_buf;
|
|
struct xfs_dinode;
|
|
struct xfs_mount;
|
|
struct xfs_trans;
|
|
|
|
/*
|
|
* Allocation parameters for inode allocation.
|
|
*/
|
|
#define XFS_IALLOC_INODES(mp) (mp)->m_ialloc_inos
|
|
#define XFS_IALLOC_BLOCKS(mp) (mp)->m_ialloc_blks
|
|
|
|
/*
|
|
* For small block file systems, move inodes in clusters of this size.
|
|
* When we don't have a lot of memory, however, we go a bit smaller
|
|
* to reduce the number of AGI and ialloc btree blocks we need to keep
|
|
* around for xfs_dilocate(). We choose which one to use in
|
|
* xfs_mount_int().
|
|
*/
|
|
#define XFS_INODE_BIG_CLUSTER_SIZE 8192
|
|
#define XFS_INODE_SMALL_CLUSTER_SIZE 4096
|
|
#define XFS_INODE_CLUSTER_SIZE(mp) (mp)->m_inode_cluster_size
|
|
|
|
/*
|
|
* Make an inode pointer out of the buffer/offset.
|
|
*/
|
|
#define XFS_MAKE_IPTR(mp,b,o) xfs_make_iptr(mp,b,o)
|
|
static inline struct xfs_dinode *
|
|
xfs_make_iptr(struct xfs_mount *mp, struct xfs_buf *b, int o)
|
|
{
|
|
return (xfs_dinode_t *)
|
|
(xfs_buf_offset(b, o << (mp)->m_sb.sb_inodelog));
|
|
}
|
|
|
|
/*
|
|
* Find a free (set) bit in the inode bitmask.
|
|
*/
|
|
#define XFS_IALLOC_FIND_FREE(fp) xfs_ialloc_find_free(fp)
|
|
static inline int xfs_ialloc_find_free(xfs_inofree_t *fp)
|
|
{
|
|
return xfs_lowbit64(*fp);
|
|
}
|
|
|
|
|
|
#ifdef __KERNEL__
|
|
/*
|
|
* Allocate an inode on disk.
|
|
* Mode is used to tell whether the new inode will need space, and whether
|
|
* it is a directory.
|
|
*
|
|
* To work within the constraint of one allocation per transaction,
|
|
* xfs_dialloc() is designed to be called twice if it has to do an
|
|
* allocation to make more free inodes. If an inode is
|
|
* available without an allocation, agbp would be set to the current
|
|
* agbp and alloc_done set to false.
|
|
* If an allocation needed to be done, agbp would be set to the
|
|
* inode header of the allocation group and alloc_done set to true.
|
|
* The caller should then commit the current transaction and allocate a new
|
|
* transaction. xfs_dialloc() should then be called again with
|
|
* the agbp value returned from the previous call.
|
|
*
|
|
* Once we successfully pick an inode its number is returned and the
|
|
* on-disk data structures are updated. The inode itself is not read
|
|
* in, since doing so would break ordering constraints with xfs_reclaim.
|
|
*
|
|
* *agbp should be set to NULL on the first call, *alloc_done set to FALSE.
|
|
*/
|
|
int /* error */
|
|
xfs_dialloc(
|
|
struct xfs_trans *tp, /* transaction pointer */
|
|
xfs_ino_t parent, /* parent inode (directory) */
|
|
mode_t mode, /* mode bits for new inode */
|
|
int okalloc, /* ok to allocate more space */
|
|
struct xfs_buf **agbp, /* buf for a.g. inode header */
|
|
boolean_t *alloc_done, /* an allocation was done to replenish
|
|
the free inodes */
|
|
xfs_ino_t *inop); /* inode number allocated */
|
|
|
|
/*
|
|
* Free disk inode. Carefully avoids touching the incore inode, all
|
|
* manipulations incore are the caller's responsibility.
|
|
* The on-disk inode is not changed by this operation, only the
|
|
* btree (free inode mask) is changed.
|
|
*/
|
|
int /* error */
|
|
xfs_difree(
|
|
struct xfs_trans *tp, /* transaction pointer */
|
|
xfs_ino_t inode, /* inode to be freed */
|
|
struct xfs_bmap_free *flist, /* extents to free */
|
|
int *delete, /* set if inode cluster was deleted */
|
|
xfs_ino_t *first_ino); /* first inode in deleted cluster */
|
|
|
|
/*
|
|
* Return the location of the inode in bno/len/off,
|
|
* for mapping it into a buffer.
|
|
*/
|
|
int
|
|
xfs_dilocate(
|
|
struct xfs_mount *mp, /* file system mount structure */
|
|
struct xfs_trans *tp, /* transaction pointer */
|
|
xfs_ino_t ino, /* inode to locate */
|
|
xfs_fsblock_t *bno, /* output: block containing inode */
|
|
int *len, /* output: num blocks in cluster*/
|
|
int *off, /* output: index in block of inode */
|
|
uint flags); /* flags for inode btree lookup */
|
|
|
|
/*
|
|
* Compute and fill in value of m_in_maxlevels.
|
|
*/
|
|
void
|
|
xfs_ialloc_compute_maxlevels(
|
|
struct xfs_mount *mp); /* file system mount structure */
|
|
|
|
/*
|
|
* Log specified fields for the ag hdr (inode section)
|
|
*/
|
|
void
|
|
xfs_ialloc_log_agi(
|
|
struct xfs_trans *tp, /* transaction pointer */
|
|
struct xfs_buf *bp, /* allocation group header buffer */
|
|
int fields); /* bitmask of fields to log */
|
|
|
|
/*
|
|
* Read in the allocation group header (inode allocation section)
|
|
*/
|
|
int /* error */
|
|
xfs_ialloc_read_agi(
|
|
struct xfs_mount *mp, /* file system mount structure */
|
|
struct xfs_trans *tp, /* transaction pointer */
|
|
xfs_agnumber_t agno, /* allocation group number */
|
|
struct xfs_buf **bpp); /* allocation group hdr buf */
|
|
|
|
/*
|
|
* Read in the allocation group header to initialise the per-ag data
|
|
* in the mount structure
|
|
*/
|
|
int
|
|
xfs_ialloc_pagi_init(
|
|
struct xfs_mount *mp, /* file system mount structure */
|
|
struct xfs_trans *tp, /* transaction pointer */
|
|
xfs_agnumber_t agno); /* allocation group number */
|
|
|
|
#endif /* __KERNEL__ */
|
|
|
|
#endif /* __XFS_IALLOC_H__ */
|