xfs
[Top] [All Lists]

[PATCH 0/6] XFS: Track reclaimable inodes in inode cache

To: xfs@xxxxxxxxxxx
Subject: [PATCH 0/6] XFS: Track reclaimable inodes in inode cache
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Wed, 8 Oct 2008 08:54:34 +1100
Move the tracking of reclaimable inodes into the inode radix trees.
This currently does not replace the reclaim flags in the inode,
rather it allows traversal of all reclaimable inodes by walking the
per-AG inode radix trees without needing a separate list. This
enables us to remove a struct list_head from the struct xfs_inode
and the xfs_mount, as well as a filesystem global lock which also
has the benefit of removing a point of serialisation during inode
reclaim.

Like the matching sync code, this also allows reclaim of inodes
in ascending inode numbers which substantially improves I/O
patterns during reclaim driven inode flushing.

Version 2:
o clean up series based on review comments
o added bug fix to prevent looping when we overflow the AG
  inode number index (fixes sync code as well).

<Prev in Thread] Current Thread [Next in Thread>