Return to xfs_trans_extfree.c CVS log

Up to [Development] / xfs-linux

Radix tree based inode caching

One of the perpetual scaling problems XFS has is indexing
it's incore inodes. We currently uses hashes and the default
hash sizes chosen can only ever be a tradeoff between memory
consumption and the maximum realistic size of the cache.

As a result, anyone who has millions of inodes cached on a
filesystem needs to tunes the size of the cache via the ihashsize
mount option to allow decent scalability with inode cache
operations.

A further problem is the separate inode cluster hash, whose size is
based on the ihashsize but is smaller, and so under certain
conditions (sparse cluster cache population) this can become
a limitation long before the inode hash is causing issues.

The following patchset removes the inode hash and cluster hash
and replaces them with radix trees to avoid the scalability
limitations of the hashes. It also reduces the size of the
inodes by 3 pointers....
Merge of xfs-linux-melb:xfs-kern:29481a by kenmcd.

  Convert xfs inode caches from hashes to radix trees.

/*
 * 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
 */
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_types.h"
#include "xfs_log.h"
#include "xfs_inum.h"
#include "xfs_trans.h"
#include "xfs_sb.h"
#include "xfs_ag.h"
#include "xfs_dmapi.h"
#include "xfs_mount.h"
#include "xfs_trans_priv.h"
#include "xfs_extfree_item.h"

/*
 * This routine is called to allocate an "extent free intention"
 * log item that will hold nextents worth of extents.  The
 * caller must use all nextents extents, because we are not
 * flexible about this at all.
 */
xfs_efi_log_item_t *
xfs_trans_get_efi(xfs_trans_t	*tp,
		  uint		nextents)
{
	xfs_efi_log_item_t	*efip;

	ASSERT(tp != NULL);
	ASSERT(nextents > 0);

	efip = xfs_efi_init(tp->t_mountp, nextents);
	ASSERT(efip != NULL);

	/*
	 * Get a log_item_desc to point at the new item.
	 */
	(void) xfs_trans_add_item(tp, (xfs_log_item_t*)efip);

	return (efip);
}

/*
 * This routine is called to indicate that the described
 * extent is to be logged as needing to be freed.  It should
 * be called once for each extent to be freed.
 */
void
xfs_trans_log_efi_extent(xfs_trans_t		*tp,
			 xfs_efi_log_item_t	*efip,
			 xfs_fsblock_t		start_block,
			 xfs_extlen_t		ext_len)
{
	xfs_log_item_desc_t	*lidp;
	uint			next_extent;
	xfs_extent_t		*extp;

	lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)efip);
	ASSERT(lidp != NULL);

	tp->t_flags |= XFS_TRANS_DIRTY;
	lidp->lid_flags |= XFS_LID_DIRTY;

	next_extent = efip->efi_next_extent;
	ASSERT(next_extent < efip->efi_format.efi_nextents);
	extp = &(efip->efi_format.efi_extents[next_extent]);
	extp->ext_start = start_block;
	extp->ext_len = ext_len;
	efip->efi_next_extent++;
}


/*
 * This routine is called to allocate an "extent free done"
 * log item that will hold nextents worth of extents.  The
 * caller must use all nextents extents, because we are not
 * flexible about this at all.
 */
xfs_efd_log_item_t *
xfs_trans_get_efd(xfs_trans_t		*tp,
		  xfs_efi_log_item_t	*efip,
		  uint			nextents)
{
	xfs_efd_log_item_t	*efdp;

	ASSERT(tp != NULL);
	ASSERT(nextents > 0);

	efdp = xfs_efd_init(tp->t_mountp, efip, nextents);
	ASSERT(efdp != NULL);

	/*
	 * Get a log_item_desc to point at the new item.
	 */
	(void) xfs_trans_add_item(tp, (xfs_log_item_t*)efdp);

	return (efdp);
}

/*
 * This routine is called to indicate that the described
 * extent is to be logged as having been freed.  It should
 * be called once for each extent freed.
 */
void
xfs_trans_log_efd_extent(xfs_trans_t		*tp,
			 xfs_efd_log_item_t	*efdp,
			 xfs_fsblock_t		start_block,
			 xfs_extlen_t		ext_len)
{
	xfs_log_item_desc_t	*lidp;
	uint			next_extent;
	xfs_extent_t		*extp;

	lidp = xfs_trans_find_item(tp, (xfs_log_item_t*)efdp);
	ASSERT(lidp != NULL);

	tp->t_flags |= XFS_TRANS_DIRTY;
	lidp->lid_flags |= XFS_LID_DIRTY;

	next_extent = efdp->efd_next_extent;
	ASSERT(next_extent < efdp->efd_format.efd_nextents);
	extp = &(efdp->efd_format.efd_extents[next_extent]);
	extp->ext_start = start_block;
	extp->ext_len = ext_len;
	efdp->efd_next_extent++;
}