1. 23 2月, 2006 1 次提交
  2. 21 2月, 2006 1 次提交
  3. 13 2月, 2006 1 次提交
    • S
      [GFS2] Fix for root inode ref count bug · 7359a19c
      Steven Whitehouse 提交于
      Umount is now working correctly again. The bug was due to
      not getting an extra ref count when mounting the fs. We
      should have bumped it by two (once for the internal pointer
      to the root inode from the super block and once for the
      inode hanging off the dcache entry for root).
      
      Also this patch tidys up the code dealing with looking up
      and creating inodes. We now pass Linux inodes (with gfs2_inodes
      attached) rather than the other way around and this reduces code
      duplication in various places.
      Signed-off-by: NSteven Whitehouse <swhiteho@redhat.com>
      7359a19c
  4. 31 1月, 2006 1 次提交
    • S
      [GFS2] Add gfs2_internal_read() · f42faf4f
      Steven Whitehouse 提交于
      Add the new external read function. Its temporarily in jdata.c
      even though the protoype is in ops_file.h - this will change
      shortly. The current implementation will change to a page cache
      one when that happens.
      
      In order to effect the above changes, the various internal inodes
      now have Linux inodes attached to them. We keep the references to
      the Linux inodes, rather than the gfs2_inodes in the super block.
      
      In order to get everything to work correctly I've had to reorder
      the init sequence on mount (which I should probably have done
      earlier when .gfs2_admin was made visible).
      Signed-off-by: NSteven Whitehouse <swhiteho@redhat.com>
      f42faf4f
  5. 18 1月, 2006 1 次提交
    • S
      [GFS2] Remove gfs2_databuf in favour of gfs2_bufdata structure · 64fb4eb7
      Steven Whitehouse 提交于
      Removing the gfs2_databuf structure and using gfs2_bufdata instead
      is a step towards allowing journaling of data without requiring the
      metadata header on each journaled block. The idea is to merge the
      code paths for ordered data with that of journaled data, with the
      log operations in lops.c tacking account of the different types of
      buffers as they are presented to it. Largely the code path for
      metadata will be similar too, but obviously through a different set
      of log operations.
      Signed-off-by: NSteven Whitehouse <swhiteho@redhat.com>
      64fb4eb7
  6. 17 1月, 2006 1 次提交