Revision 1:
  -mm1

Revision 2:
  Fix 64/32 bit ioctl problems.

Revision 3:
  Check the uptodate flag in sub-bios to see if there was an error.
  [Mike Christie]

Revision 4:
  Handle interrupts within suspend.

Revision 5:
  dm.c: Use waitqueue_active() to save a couple of cycles.
  
  I couldn't think of a reason why event_callback() was using
  wake_up_interruptible() rather than plain old wake_up(), so changed
  it.

Revision 6:
  dm_resume: dm_table_unplug_all() was being called on md->map, rather
  than the 'map' we carefully took a reference on earlier on in the
  function.

Revision 7:
  Log an error if the target has unknown target type, or zero length.

Revision 8:
  Documentation for the linear target.  The last 2 scripts still need
  testing.

Revision 9:
  Correctly align the dm_target_spec structures during retrieve_status().
  
  This patch applies to both 2.4 and 2.6 kernels.

Revision 10:
  Clarify the comment regarding the "next" field in struct dm_target_spec. The
  "next" field has different behavior if you're performing a DM_TABLE_STATUS
  command than it does if you're performing a DM_TABLE_LOAD command.
  
  See populate_table() and retrieve_status() in drivers/md/dm-ioctl.c for more
  details on how this field is used.

Revision 11:
  VFS lock patch.  [Chris Mason]

Revision 12:
  Lock the filesystem while a device is suspended.  [Kevin Corry, Joe Thornber]

Revision 13:
  Reverse out the VFS lock patch.

Revision 14:
  New VFS-lock patch (from 4/1/2004).
  - Replaces patch 8 from 2.6.5-rc1-udm1.

Revision 15:
  dm.c: changes to use new VFS lock patch.  [Kevin Corry]

Revision 16:
  Fix XFS to work with the new VFS-lock patch.

Revision 17:
  Multipath target

Revision 18:
  Multipath: move the path failure decision to the path                                 
  selector. Some ps's may want to select paths that have a couple                                   
  failures over one that has N failures. If a ps has to track this info                             
  there didn't seem to be a compelling reason to also do it in                                      
  dm-mpath.c. Well code duplication and sharing is a good reason to do                              
  it in dm-mpath. It would be nice if I could just add an init function                             
  to any ps and leach off other people's work :) but is that too modular?                           
                                                                                                    
  Additionally, if a ps fails to activate a path or cannot use it                                   
  during initialization it may want to fail it.  [Mike Christie]

Revision 19:
  Multipath: exports the register and unregister ps functions.
  [Mike Christie]

Revision 20:
  Multipath: makes priority group / path-selector initialization
  possible. Remapping failed bios from end_io was not nice becuase you
  couldn't choose a path until you knew a group was activated and what
  paths in that group were still functional.  Plus, incoming IO needed
  to be mapped while failures were coming in.
  
  I guess you could just make map_io sleep or do some internal queueing.
  
  I choose the former, so I had to move the remap map_io call to
  dispatch_failed_ios and add a get_mapinfo accessor function.
  
  I think queueing incoming io while you finish the failed may be better,
  but targets calling dm's queueing mechanism didn't seem right, and
  reimplementing it didn't either. Well, I do not know if it is better
  or not. While paths are failing performance is going to take a hit
  so io ordering may only be a problem if you are concerned with
  journaling or barrier type stuff (but as we end up sending IO down
  multiple queues we cannot gaurantee ordering on the normal IO code
  path anyways).  [Mike Christie]

Revision 21:
  Multipath: adds another accessor so path-selectors can access a path's
  bdev. This is necessary for devices which need to send commands down
  to the device. (see rdac-test-support.patch)  [Mike Christie]

Revision 22:
  Multipath: dispatch_failed_ios can sleep. It should not                                  
  disrupt other users of the default work queue. Under heavy load                                   
  or group/ps initializations on a single processor machine, this                                   
  conflict becomes painful.  [Mike Christie]

Revision 23:
  Multipath: rdac path selector.  This is just an example of what needs
  to could be done by the ps to send a failover command to the
  device. This example is just sending a TUR, and I just bascially
  ripped off the rr ps's other functions. The point is that I have to
  inject a command into a queue using blk_execute_rq or elv_add_request.
  [Mike Christie]

Revision 24:
  Multipath: rename rdac.c to dm-ps-rdac.c

Revision 25:
  dm-io

Revision 26:
  Eliminate another compiler warning on ppc64:
  
    CC      drivers/md/dm-io.o
  drivers/md/dm-io.c: In function `vm_dp_init':
  drivers/md/dm-io.c:441: warning: cast from pointer to integer of different size

Revision 27:
  BIO_RW_SYNC is a shift-value.

Revision 28:
  kcopyd

Revision 29:
  snapshot target.

Revision 30:
  mirror target

Revision 31:
  PPC-64 generates this somewhat cryptic compiler warning:
  
    CC      drivers/md/dm-log.o
  drivers/md/dm-log.c: In function `count_bits':
  drivers/md/dm-log.c:433: warning: passing arg 1 of `find_next_bit' discards qualifiers from pointer target type
  drivers/md/dm-log.c:436: warning: passing arg 1 of `find_next_bit' discards qualifiers from pointer target type
  
  Removing the "const" qualifier from dm-log.c::count_bits() eliminates the
  warning.

Revision 32:
  Flakey target

Revision 33:
  dm-zero